From 98ff83bc772c0d4423f2202e7b71cd28bc7927bb Mon Sep 17 00:00:00 2001 From: marinkaz Date: Sun, 18 Aug 2024 12:14:17 -0400 Subject: [PATCH] Updates --- .../08/15/PINNACLENews}/index.html | 69 +- DMAI/index.html | 58 +- bio/index.html | 2 +- data/index.html | 58 +- docs/marinka_zitnik_CV.pdf | Bin 91426 -> 0 bytes feed.xml | 2 +- index.html | 58 +- jobs/index.html | 66 +- meetings/index.html | 58 +- news/index.html | 56 +- news/page2/index.html | 56 +- news/page3/index.html | 28 + people/index.html | 86 ++- postdoc-ML/index.html | 60 +- postdoc-TDC/index.html | 58 +- postdoc-biomedicalAI-MGB/index.html | 58 +- postdoc-cancerTxAI/index.html | 58 +- postdoc-medicalAI/index.html | 60 +- products/ada_fang/index.html | 4 +- products/alejandro_velez_arce/index.html | 4 +- products/andrew_shen/index.html | 4 +- products/ayush_noori/index.html | 4 +- products/george_dasoulas/index.html | 4 +- products/intae_moon/index.html | 4 +- products/jason_poulos/index.html | 4 +- products/jonathan_schwarz/index.html | 195 ------ products/kevin_li/index.html | 4 +- products/kushan_weerakoon/index.html | 195 ------ products/marinka_zitnik/index.html | 4 +- products/michelle_dai/index.html | 8 +- products/michelle_li/index.html | 4 +- products/owen_queen/index.html | 4 +- products/pengwei_sui/index.html | 4 +- products/richard_zhu/index.html | 4 +- products/robert_calef/index.html | 4 +- products/ruth_johnson/index.html | 4 +- products/shanghua_gao/index.html | 4 +- products/valentina_giunchiglia/index.html | 4 +- products/wanxiang_shen/index.html | 4 +- products/xiang_lin/index.html | 4 +- products/xiaorui_su/index.html | 4 +- products/yasha_ektefaie/index.html | 4 +- products/yepeng_huang/index.html | 4 +- products/zaixi_zhang/index.html | 4 +- products/zhenglun_kong/index.html | 4 +- projects/G-Meta/index.html | 58 +- projects/GNNDelete/index.html | 58 +- projects/GNNGuard/index.html | 58 +- projects/GraphXAI/index.html | 58 +- projects/Milieu/index.html | 58 +- projects/NIFTY/index.html | 58 +- projects/PDGrapher/index.html | 58 +- projects/PINNACLE/index.html | 66 +- projects/PocketGen/index.html | 58 +- projects/PrimeKG/index.html | 58 +- projects/REMAP/index.html | 58 +- projects/Raincoat/index.html | 58 +- projects/Raindrop/index.html | 58 +- projects/SHEPHERD/index.html | 58 +- projects/SIPT/index.html | 58 +- projects/SPECTRA/index.html | 58 +- projects/SubGNN/index.html | 58 +- projects/TF-C/index.html | 58 +- projects/TimeX/index.html | 58 +- projects/TxGNN/index.html | 58 +- projects/UniTS/index.html | 58 +- projects/metapaths/index.html | 58 +- projects/patient-safety/index.html | 58 +- projects/scCIPHER/index.html | 58 +- publications/index.html | 58 +- pubs.json | 40 +- research/index.html | 607 +++++++++++++++++- sitemap.xml | 58 +- software/index.html | 58 +- talks/index.html | 5 + 75 files changed, 1890 insertions(+), 1721 deletions(-) rename {products/ivy_liang => 2024/08/15/PINNACLENews}/index.html (70%) delete mode 100644 docs/marinka_zitnik_CV.pdf delete mode 100644 products/jonathan_schwarz/index.html delete mode 100644 products/kushan_weerakoon/index.html diff --git a/products/ivy_liang/index.html b/2024/08/15/PINNACLENews/index.html similarity index 70% rename from products/ivy_liang/index.html rename to 2024/08/15/PINNACLENews/index.html index be425c64..c3dacfe7 100644 --- a/products/ivy_liang/index.html +++ b/2024/08/15/PINNACLENews/index.html @@ -4,33 +4,31 @@ - <a href="#">Ivy Liang</a> - Zitnik Lab + How Proteins Behave in Context - Zitnik Lab -Ivy Liang | Zitnik Lab +How Proteins Behave in Context | Zitnik Lab - + - - - - + + + + - - + - - + +{"url":"https://zitniklab.hms.harvard.edu/2024/08/15/PINNACLENews/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/2024/08/15/PINNACLENews/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"How Proteins Behave in Context","dateModified":"2024-08-15T00:00:00-04:00","description":"Harvard Medicine News on our new AI tool that captures how proteins behave in context. Kempner Institute on how context matters for foundation models in biology.","datePublished":"2024-08-15T00:00:00-04:00","@type":"BlogPosting","@context":"https://schema.org"}
diff --git a/DMAI/index.html b/DMAI/index.html index b13ca9ee..a19f82dc 100644 --- a/DMAI/index.html +++ b/DMAI/index.html @@ -202,6 +202,34 @@

Coordinator

+
+
+ +
+ +

Aug 2024:   How Proteins Behave in Context

+ + +
+ +
+ + + +

Harvard Medicine News on our new AI tool that captures how proteins behave in context. Kempner Institute on how context matters for foundation models in biology.

+ + + + + +
+ + + +
+
+
@@ -746,36 +774,6 @@

Coordinator

-
-
- -
- -

Oct 2023:   Structure-Based Drug Design

- - -
- -
- - - -

Geometric deep learning has emerged as a valuable tool for structure-based drug design, to generate and refine biomolecules by leveraging detailed three-dimensional geometric and molecular interaction information.

- - - - - - -
- - - -
-
-
diff --git a/bio/index.html b/bio/index.html index 49ea887d..2bd3d476 100644 --- a/bio/index.html +++ b/bio/index.html @@ -202,7 +202,7 @@

Dr. Zitnik co-founded Therapeutics Data Commons and is the faculty lead of the AI4Science initiative.

-

Dr. Zitnik is the recipient of the 2022 Young Mentor Award at Harvard Medical School.

+

Dr. Zitnik is the recipient of the 2022 Young Mentor Award at Harvard Medical School.

diff --git a/data/index.html b/data/index.html index 049a919e..043bd50d 100644 --- a/data/index.html +++ b/data/index.html @@ -741,6 +741,34 @@
+
+
+ +
+ +

Aug 2024:   How Proteins Behave in Context

+ + +
+ +
+ + + +

Harvard Medicine News on our new AI tool that captures how proteins behave in context. Kempner Institute on how context matters for foundation models in biology.

+ + + + + +
+ + + +
+
+
@@ -1285,36 +1313,6 @@
-
-
- -
- -

Oct 2023:   Structure-Based Drug Design

- - -
- -
- - - -

Geometric deep learning has emerged as a valuable tool for structure-based drug design, to generate and refine biomolecules by leveraging detailed three-dimensional geometric and molecular interaction information.

- - - - - - -
- - - -
-
-
diff --git a/docs/marinka_zitnik_CV.pdf b/docs/marinka_zitnik_CV.pdf deleted file mode 100644 index 7ce3a7593290a290b30ba85ff9b118f1b75541ac..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 91426 zcma&NL$fdptY*7y+qP}nwr$(CZQHhOyxX>ItG`=y2dAogaAx@dsZ`dJ6;cHeF|bqF-sd4Qzv>c8$%aU5mRG(6H|H_Q#*4P3j$^~c1}J% zC}$TZQ$t%QkIh+~$++VV#NIFJM{ucOU8ID6g99@FuajasF!)VW%x4Dg27iBpB*umg zic%{(#xPz^Rnn8B7%?7=RrG(I=h0FA&!%h-{H`D0?%y|S?;o{)oExkve}7-~!Qp>D zH;-p;)AIf2(JM<$6z-n>dNK;i4e1P_q&mBIe;rh&+hytROX2U0_m1uFj`uS=4`;^S z%xgHmA922SY>-vo*=vb#`qcP^^j= zZqhrdBd2a?lXD=n?8iqLV*;VDganIIiBANLo ztUYkT#b1TC<@6)eliSkhL~c3a-DG51ByEHcQTgpN-Al;hN+@^D`kJ-Nra5^pM#a1g zm{H^FV0WUTw68j{grX0IHFoLhFlg*e0713PC zoN3i__yIvgtF6<#O2Ws_7(KoN%AfU)d5vs8$?(79_GkfeOXlgaYdJUzl*)dPOscSXXR&zzVI7ivj5Hm% z_XELq9jS>K-YCWBSeWc{k5BY$Y3rlg?NwDY;(X|#VmqXfCDt5(ATNPiYq|InMn{&i zG`~QgsfPr+?z0Ip=GR85x3X-2iHh(vMf^TTB}=1k;Q`B6vY}!3&B}N_9+!Zqq~fy7 zoWPk1I;|h%G>EK(p}nOUbCO}?o}|=>KAXCLtuhAXJbS5LXRpzfXrbE;`bll9@qMpy4CP8QLc+y&mt$v-vaHY=-=&V7`uvlR?DJ zA89z~xPd?1GV23n)VIT0J25Q%wBQb4r?zS*3{DIGZMbl7DGc+kDuu+d+twe0p`-`A z9l4g$0OP%HKCuW1shf=_gP(ABL!-M24(_B;N0dN2>LRNxB5FzC?Wz(rsWFjyG8kCw zPam$h@t8Isn3&l49E$FfJ;Dg4k@|&eW*&6;1na#>h#<|c;UVWd9}hQ`Z(mGnia{|v zhd_lTqgjn^H9bYi$(}0gn*Wsv$egH#sv-Qe3Z9y%EUVVdvqN_Wf*@VgnTOZUkrCs0 zx)U85)~3&pVdr~J#f2QVO+B{G9&^z;=!yRB)mS%Yt4ZAQ;O5f5sX35A&qwz@EhqL+SRM`0$#)c?aAsq{p(604q3iX@p#D@TyXi{Q|dr<|T^ z5+YD&;$kksuyge$xzF_0!;ODcx|9v(6=B2SVx9aKB~lQT$=?fDW|5>q6M znbW|y-uTDe!nid+p@PyGSUaSub3*TUm<+&bug<_>T6oy%^_osm4HOMKEbj@js;1G5 zWExYMfowt>v;`drWZ#27HXv6%jDe+(;=vB}%oN#*@1=Q{g`>uo>JIycPxLKEUBe?0 z1j*xhAl%7@L`3CtMu4JuT8;K8>0Cnc<)oysER}o&f~8Rz){fK~eCJ^t^w}KCq;di< zzDre|*>xKND&yfgr>1h!sfA=o#)E!t*9Z$k6yuT?H0R;UpLJs+Dz^?xI!b!At z{i{+yt(X-%_HXStQ88XQa{#1pyky!myMA3CxP?bVp${#Ce*9X;thSyvf;GtO-xUH_ z(%rKDIFNBjr|8u%hzce0-D?hBVu4;@8+I){`BhUoh?R+OEky9eEPLuM1w2 z>9Jv-am2}sG7Iy>+vz9wYaCMjEvkp(Igu~g4i~MMKWo?mu=xOcrcE>ya}i~2;paX& zevqJZQ5T^+&MlM54Dz)a06Tx-v4CnT$1zI1_-??$5TLBJ5?HwMOPX*KGAg0LR%TSf z#UWX7IAJ$hY0d*xx1*8Tcp)hlM+2j7$mgeiyQ)H(hcrI5x1VT|djGxEo>}4-*%}^pGfTQ4v(A!9f|F z*5vNA&-uiC%3)TOaKe-=FkPEcV9(F1_nf*V+#RQV@^n?Z2!l;#S*aSdgj;mJqkm24KEN5c&o_?g1X7nvKGM}qu(>6zRDXB$vDTuIg(MH>(d$|zD| z1V(^El`P`%4yLxI@3~1oh~d+!;8IA(>5>lt<3&i^i&U6*j+Qh(KZvtd)VDlC`?zA$ z9M!JCdynxHEE0g0Di<3RG)!g-O9tY;7F6Yu^)<%s=f}>mbEzL1?EJO01K)Ho45^R} z0d{ved)stR6Q(rst<8gcc^2aV*7O0fMd-gu%Y$BxT~bx1>+)ol2zTn>PwY{WsErW;gv8rjBGDiFQP;H2c$6gSy<|; z25Csk4pGzxgB2wMpkTx%pabD|yo3!1Juy8{W7(A{ZghCIVzr=|DYkvh-5Jt7JbWWDGhfm6_}^f-aD)fz5SOke$otjW)b19hvZj<8jbT zxr*e~c~z7x$Rg^U=D%=Jx3!KsXJHok%LTgT9VRuL#|MP5?JjWp9(a_yAf>=6r#RkB zRirLfT+QrY6kSgSCr#ERa$cLfEA>@WHJ} zu)j*@<(e;jE)m4+T1! z%lfc*d;orD9)cyRE?#}w((^Clwnk^Ak*2T{b}VP9%@uOCM2ep|GI^Z@J`cfgv&J9B zr*6C0RjsY7_-|QLxcTL_Jet}nlbn=(V6dSZOEa_|cLnUm&y)!vb39rWg>aVFN1 zx2w(G*U|Sn>Gkw96tVZ}VXJGGk>vyg;rWW$Q*H1+W=ri29_*JOPttDaPT0G{ zgfyKT+hL;}d}msi5A5@8ZUmkWTbzY6IKm~<|6>xa@lQcFBraMJ&z&DAo@b?xk0&}& zH<`+I&v^yf6@v-2W~Q6L8K_l9w9DP~CNw0t+5a-T)AUw9aRQfw@25!)af zqH|M<7Fe%>E}-Bsb@8K7@InrBV&ewL^}H3m>ZiXya870@>o%if~<3S!YZ(z+MJ zB*ToWQGqe(jT$=W7v)IVM2B00635F)P;d)V0LqyM+aXx!kXv1oHI01@2fi7H)f6%K z;_j@+mlb^$Hb}q}7{?;BT}5$BH0VJ3B8*73 zXC|CEwaY#CJQ{(>FU_V=$-+IvU|62M3D6qGsf-ICKN7=O)k zlGy!gAcR%x8b)H?Dt<-6FIGX$WeSLe!n$Jb7VOhFf#Zk8Xec!sJzvR4|7 zk8$5VF#m;k{V9#HTR$IX`c|x()Y8=};yWxE`N$xDbL5lCCd)D79B+?&*0Y^jTRkjo z$$RfSqP5I=AgV#Rt5R2#>)HlflvL)C;y9%i2N~=~!ihk*!xN@w2KSKOv z;*F!Hcub{(Br%e)j^_Xi9O`yZNHujcLjHw|n761hh45 zqlu*fo3=Tm$<-8F5}WQyyYs0Rss#@6OKkIvSHD5@4ItJV^+t&kxsz$06c{Y@E2oQ; zEx|(Ep!EGqqB|sed)qSo*Qz<-^H%VX_Ch?+Y)IX2(M}Zi*<3&$3fcDX8|jE0PuSHH zjmQ)H!d)Pz%1$T``$NJeqc~-w+YsdS0U{U)eYXqkHqct}e!Z!y?aZL+Zs-6!hcQ48 zA^Cy2CvKoq9j`V9>8wajr#$tb314GkuJsb%n5s4V&ovQ7_+41JXw47}D~0ut;rd6y zXM}Mx!?pNjt+X zGP?u{n72~wj?3-s)gSl{PVn&v0MQ2hQts;*cxB*mLf{DC&{|OsWg$OET^5 z_o8vMi)asH6f9mu11D4R1$AM`;4a-?qAySX-waWJ7Ae1Ucv#=TlGBn*u=cMzLlk z)Fh{bHs6QAS4*U0Zl!3m5oEhj7Gv6cBbFC25NY$AohMt|?3jpSH47#?<}*4yy!ifA zqa(-~JYfp=*@?)7^`O7rlosKOGHdI(o537dgmb`%#dr@^Ml9^c0q(-Slv4wXjEb#n zOEC%zdZHTBw-no8HrtGYtP-((d`+F`Np+M0=()|hTx(-xTdaS_;qOBWUjirJleI&7t}y?eZgG6Ep<3&Ort( z4MQXrgpg|+21do(=e{S41{jH%ybm)JXX=%S&^1xwmNqLN3KX|?=C{V7Wvw7)E5*7R zLuN`LV!E!(s#SsF!U2@mg+p&w;;x+9dplbae4_fq=uE4me?eEHirVHwQ83B9BagwS zLxI6jI3ztE8{}6S8jG+1zHg#_fefs8QB6rj-#&6^_DapjZd0 z-83_iKb9K-gIzC^_zZ(vX?xC(B~NQaat&J|PA&)RlFEL%nXb%SuM;czzmg!;@KiGF zncn2QO2U6vm08Q2vTU+oC4)Fs^rD(ms^Rx&Q||~@6wD^`GSR8z31QkR;kZ@B(%W-v z_>~(gN;neS34wz*-}vhrk-6uOQpUSOaQHwpj9}{cvP$o)uX*zbxPqn;hRNu_N8W2e{GNyc_k&@HXO!d#q>!ukEp<}emTeQDht!-BKAKeExyvv zJ`hqt-AeuNIpFA}ZDcs-m^PzCPI%s3Y?Zn(O+y7fN+1(P;BB}n5>7m_kg1K#@G2x} zdd{%r@6)6o|J0_P{hG%VDvD8@5dN^zkehhcTJjndF-k4EbrD2SCn7bu-LFE4%HNWa zsZGD{f%eT|W&6ljgEFAb@?a=&zDptydu@P0bxh;`%| zHquth>Q2zy5vP$prlu^@>=N`7o}?%49{b+1;4G42T2F)&=TO#=#DYviQ}CvkINmT3 zo}WrqymQ9GEuj5Jg=)SG@suI#&Z+uRc#?#~Xlr}IlnpPS+jn|Vv|br$Fdg)R`L7p~ z&5)diyrZnFSBA38*w6nmvmT`^0k!jx(B(Qo8^Jnn7taPTj6cQ`8f3-lzF);n7+ABig2Y*s}Y}4f=aqkXvCEynK-4ccH$$T|RM`~20 zq^e9y4d>}}XHi4dV&l8n6O?9$UkKezy;l1EI)SFDn5$wlXrhgj0i{_%YXTd6bUc#u zXk{K@Y=Z2svjA06O_hPSjz=9Gtrncx$tKmM6zQ|g(2Ln$(mopy^ITlkny|7rTdA3k zI)fC#Q(di7C#3MNnby9$A< zN7bKUDMd8L#|vbGxUsI_&FXa<{7yyJ?sDCK;d*8?W3uA`(8`-4RJVM(ANq`9?Dj^SU&=yyF20NEYCMW_B?6$*8 zryL*J=SYK+jn-pWV4T47Oq6BJcSoiGcrvVa&IjOvChl6aw*fnZm3zJgR%FMaE50m> zRsGZ_pyPCKOXv{UG`6kc`5Kk3Waw;4u%4Yq3;HDmEuUn5r_OxG%`1y*S8ejmO5SZs zo_8FpaT%1DL9DN-kj8~ENzr#%Jv>>gx-RY(seg3MoqE2|=tE<`5gUbP&ekT-FmsEB zi;rw(8cPa+%y?8PRnopw?{0CjWtYy`j#|SPoXfG}yVKi?E&4fP#yc8cd&IdH86N!D ztY$x$ke5OO+F0Q9G=&R&yl#5j??F-jJ(cWeKd3T~l&|%JCu9wigyNNOjTc(&qP9+m zS-S}gv*vw>`PXBuZ}FuztC^3p4DGX$Vq05fGUbS6D#u>P=s=ZX5_ry-Ib>BK}``VM)HaLi|Cg=>bfbfEYt9iZk~F4fMYrV+QWgnbA4Z4j~T zQ~3e8yP(YNKM3&nkK&e&DcA|j@^qKaUsXqQn-6P~MZ+s5m9UZ@X5OtX^19L_OBQOo z*e=bAfyTk8XrC@HSvu4&6}RdMw1<*4#K&ow^InX7*qC()(Y>bPV!qXE+yeoJBA?)K z_it5G+Np@vZO-n=4rW;Oy6f%O-FNjX<;*HB5%RU>+i{a7 z7;`q>+pzUyk$ahnC1j8fPS`jMluoW~-aKYpF1nUQhrj0ZG0ZrMkXg_9(p_?E*qWY3 zYK>F|Vk$)fv|EfS@BM;puNpOozp4v&G}SP|VUgc3izchfo)c+(9QNJM=DqHSZ@Pb? zuEUi#;>sy=Z!7y8ghbn|YF7@MMQl!b8Qu=JrkLO~sKDq}w`e(y?Zj#&MRM7Sp27?_ZRQj}#fN}M1u`3Pu2xD6u zQ2s$9cb)ie3AFoa2eJa6AlB#FD%T{Okt62%lg*uF6kkayp~9yMeKcM))FjbbDdqk) zFUn;bc@C{s^YJlz%LCe>Y&uqmZ(N6U&ungNv-`xJMG%f*NcEsHu?c_I5E^S(8&GK( z>~0RCOSY|9;E#{yfR{)_(Y3(w=|+ zz1)1pf4+Ubrt|No`~PJ9DgVB^dYOKoxbyG&eMS3AOFwS*|9JFh_Hz7`Egq`)d^vty zS@v>x|8B+e_x=2i?|Mv+b@S{>164-KQt7-urprqkd0Y4hPe2yyiUN$P8a9CgM$p(gkLiZ;Yqnxpe`&CDODWsRTj zS6w_QWO&T$r^-T`%l_59dcH);;e27d34iq%!|PaT8MWhFR=1h5aIZ{OH`JAhvF)OA z);P(mCv`LPnp^3o?6_8+1+r%^`zgpjI3rbPD%-wd3I_#NeaEYoKgd3x9&O0 zs`op-$o3e5tTRnA9U3`utGDi77FQ4J9N@UQAaf?VZ6QRwgKj{M?N?i(O$`xy)%#e= zMSe?IU(-(s_vD_wTf%xn=W|_CxZiL5jC(R&3^J04C-TL{*Lk(aWs_Tip@D0WP=Y4J zF@bZBGqF%F$-Ud&hBSvsXq*lrK%`(WMMBbJ4T{gVtUgt$ zFy9P&QiQrkFxxl}B(OkE%h*N?_=bUn@gdQ>fTr!8hEUIo4YPBZnu-Q=cY$|upPk`A zHiV?dW2T*R4Fg-D!(30!BWlU1$}vIt=-cEXZ>v_QM!}*QywSFRKMs8V=!DG8B@3YrUaF z^xarmk`+m)bIAp&untgAiKlj7Mae2eVnRz2*_qLiQkdoc*wh-w=EF6b0yD0eDG zayAZu4zzqNrt={XAVlmp980q5zM^_8pNZGEXLs+x{;ih^hF3EAb+z>xXBZ30n@iCQ z9UZ%KdL=SPHu#0X*FA=1Cfr7&sKybZ?C6DeorHYFgWL-)`CZ3}JU~Q)FM%zB$wyh7MX%ERq{;-GoElZ+*fkC}vGTDA7>CZ4-wL-o4 z*OdIOa<7lJyXIm{9~L@8$k(&ceZEuy^%)?sIkR+Mn<1M}QEA!kW^HFrgtr6}NAdKJ zh*bb3F~;(@RP;lF=5r-Gu>zx5IwM+H~g)(M+fg@UK)NblI9=#6qNvP;~ z@^rhJAVX&5n-T(V)$Rtlo0Lt?i77s_S__ z-?jz87Kd32!nWkC=Z=Q_- z3IG;8%>jt#uKFA>6se|88Vq?*)e|)l#S{o{X4_$XMTJ<28eUs~x#^52b#13b#(BO5 z0%XicVd}ly46BGhQl%{6wfN36VTIV^7G7~@Uy`$u{21r5-Eh#*^H+HZsO87DpkEt) zr?%r!r^aq>x5IdMSsLOnJj4;BE6%W-wd!k0je;1hDL@(I+PSt z&1@8EHzC%8^?u9lQ_X^&MllL6@+7r)aO53GOUBtU09U$8U#h812vGfd+|QC;3sYlX zsq4?@YBi~?Q{_OgeICVG^jgJo^aBqiyW&hLtJDFRrthj{{avl8VZ)Umyy19m(bA!s z4m$<2+K{^dTdK-WQLe(*oN$oeyW1# zFk_|EXu{6H+szb^8l%(<-oydp0&ewi7jYRbw{3tG`fW6;RQdnt4#ARmREwJDBw)^d99%? zVLFt7b_<1Ob?Qk7-l)4;O7Ov`p(~j-ptKFHkd>VUUjDNOtJ;pjX@+OglbAV*hbU2a z&R0#}TrbK9#4*qtXbW5P+`vXh6g=Z@e%E&w?nkbJDy<@!ZVjs0&^<|@b`6wpQh@~d zM;KY_`VX$o5Ijj>X+=^${kr&GwFdcwR}aZ%38h%K=?+B{R|Q277KF_PXG7CVw-_NC zZB%PS^LFbNM9MvpT&2AG8cZ#8QKIPR@yZ5~>H*`QIkR z<(6{OsjyM9=9MN@{1qnJl$i#cmkN`ON&KPGHIi#^nQ(LR~jL&{fszB zr_UoD(vqosNnvZt+E?`DR;;_$l89`_PcOFh&Kr54L3av^hh@eiv00cdN~N+wIhMHC z_9mR6qG^#Mm=SBr$<(mlnW}}Avk@G%YDm9pS^{hILnUKCxgxAI5NHEr1Y|mBIdZj_ z1IS(RyF$8$TNKeIG&)a6Y!twDEx20R7-jIGkg;W<4&`uwA(z}bXl2=(Z z_RHX)um+B#uR2;oqXG+=H-fq%N7!<%k;=jWRUyf}t_V7A&gff-jI`y~2UK%lyv4aB z4Lg`H7%tlK5#Y^7m#I)pn!I8jVZF}TsJE+7tC*=9;ByIba9$TP&aK#|r8CnAV?;yO0(xI=&0b1EfUF zZ>4EQtQ<5q(TvcxDs)BeOwzA$Vyp7v3fc)v_f=T0pxq2Fr#4hCZq-p9GEK7J3Prnw zs&EXi1L0OfOBA%Yd0f03bkvpAgUSvkh^i#~DN5HkGtm4H+QbO;(vHPhKpy@jd$nRj zCZc*_)3hj_$p#j$#KNphc=H3qPfuxsvt13j5Nh|dFQXOQG=W<9Zze9)Dq%MqreQua zegkA#iUgY5l?}*%DK1q(gggTL0=Kh{x!nyAU%SdP&EL{k_9bFt+++_r``}0fxCrM! zc?KEutMz!>^Ks3Rt!qg&GY@R45+^O;AKljs)*ZybyW|qxMQIqX#k@@|Kl>uP)BA#q zEm00%goB#7Mr#-1q0wLn_5|tUS(VXmgV`RUn?zp+`W$@7Fq<{)pwGoz&_~hJiqd7H zYc`qxbO7!Ut<6c0D8)3>;R+hgycQgm0^QDB!{=P~QzSiGHm#pf=?oyo*8G zsutvHoNk+Q0rSG5%kc#(4$OU;^7J^Y-Lr7LM!n@(rBN2YfaYcI&ZoAk(fVewC_T-f z|GFB&#H!n!Mz~q1BqF*7n>x~a@H(sUh`?jQnO=iL-3^=ptFOt3#w@Ab_pw`(_K~`nS=0f95HLQS^4Sv{$Skh)gmE!{9N{d}#r zMF@kRr#>I;sCiyPl%ZN76%iQ1lQZ_jy=n+YrGE|N&a1`8Lu5ZvCuB6r__}`zVaM! zfD0<|#Aj)6Fb-(L{}bmd)O(C6;s&I;yIfv&$9ud#>(%_P2&*%}T`HE$4rvD7m&fC9 zm=)c)f9f&{+NAG-@h?X~`vx#%<2y(-^DzI{q|}{?PS8Fn^0r_R{#XNd*dsL3V<90N ztdPnmW(M>?4=+wtXUldvf*n5$nIf2loIr75&N z<~aFjQz8#l*n0>fCWRM5;1&Z?^gqhAt#y|{vM;w8f_;%+&Eric!O>0yPR zGG&+ks?>VH3kMq&uko$dVMyy=c-iJXY|7+$n6t)NoHYs7HSa; zPt^5#7Fm6Y*My&jU)$Ag)0&9n`}o1~6EK*N(XdlJ(FG^Nq6IG&J;7U}Z#crNtubP)fAOfA*y<{CK z)%!9c(LmdD+Xe+wQ}?vcRV4*(ur&_ywj9~Y_;T7)u**I@_>zJvgMO=<+& zf}gP_8W#=t2$21&cf`j_j}YlNRZ&qMG)YDOnB$k*(|`w}@-Uko z*d1tA*5q(th*%bTb*OSU^Z^nUbwMU46Bi^foi5}nTWMJ6SW^nYsUd%dZ$hdE?CdZ+ zqc7M#I@tO!jrU}eHa2%;>_&@+a*<9UGGw7%-1#m#|AF!7F}+dacI27jIWT-X^#wcD zG%RSs{_SZgE05fd2$~2+FW;RP!&tmq)mKtc_CMGSbd2}GaUR(>jJ=@HzRq`=&24At zBYkkY7vJo1gVhe+0w8)k?&G`%H`okbcbI_ctS(>mk1`N9es_M=&lF)}X3UrcyREBT zOslS-sBdf;#yPk>-Gxske7)Cy>LZ)}yD_nZCULXS9En zwk#d*&!>-+V>2uNMwh5|o}cgcS2QCX2J$RzTl+zCZ|u#9{d}Rn+l_C)-H`QHcA-WW zuebeIGk&`{xW3`VE6K`oK88Ou`rY2PHYK;zYWl*XMO5`I?dxM$*Q!;^&+s{b8wto~ zQ!6H)m*?6vxhf(|IMvgunRxUsSo|xJPCf?7EGRF;(LNiH%F(QC%=Q^y-Inv;Tq~6f zcD`+IQ$5S&>is+SX*iaQQJePB>C-EI|8Hll8cOwv??%-H5difHbyQIi8=LeMqTaEJ zZ=7Zb*2$-Lcj$hH(2_abD|wB5J^OWMxxQEZI6IntZF$>kV=Pkd?-|?gEcoBE%??`e zc$T;?+8-C*9{;bGh*Yq>l2(4+G@5f+V}E-Tl*;KHe+7)Fo;sSHv8k;7N@LV*GGFQM z1+Gal+662*$$R4j0l?uw6!hnU-dzLU&)2&b^+9O)rMy36k$!70Yh?P+PN5jOH~4op z(x<6>FMf@ULpHr+J&gR@XPk`J*3ndQWK8;D&1aQ4?Yz2LYU0LLV{6TEoY=8bsQ? z_xL=>Y5n^TM#-OU0Pf|fjt1hEo`{Vq_CYawr~jkVVXsgq?{ZtOX78S`ZG(DYo2xWo zEi-jnR--x;4nT}P=)DCGa|uL&g9Wmi=jxOn=&~XV8zZ)D^Zq`~)l$>z0B1@cL>BtX zakXA|$GOuXXJhZ?lbfi%haOdE@!4I$nP}}9C{L!)YzI|uJKnR<-pRMDtK|H!R`!s3 z##JMyRB~BQ6*;Wuy}`O=eqg%@Kj#JZ`shzOfIzk*c0AiSW@r|5iiHL%7xFV9O9F=j zaOyyOxPm?gWeFPeeZ0iz{8t_5_=!(b1p7~7-_5)@W6KGZSAD|uZ$Z+f(vDNCv-`m^ z5zNpt8KEpJJFdl>JL5MOT5%GQV@VHLw`cVMV!<@vKMXuf%f^!SviXzl8Y$`XX7ob( zv(-+bLVDmCy&yi-Y;W-%Da=qYDjM5aNW@L000%8`ZlwH4_B%&hGr_mQDU=lKrjNOQ z%UW9wo!190G+&?`t=g`d_10`c&Y{1aO7eox7DI=zbaVj|r6h0Ql#js0Oqy20J&MC7 zEvC8zeGsLPZ~p*FK&WWuv#TQeeU};BD)*HVZfRHQes_y016h;P#HMgwfX$gGKO*s34r8!p>{OvOjRc0V$DY?^0juz8H{zW+_H=gd3ecBnTc_- ze-2Phv&GMmwpOWsA@qg_4Lp{sc+$?toQKpy`hne!-A~$w<5(S`SL-{J=?~-T;sVS4 zfO#rXKsCRkf1*bDRc_#i<&i_KpYZ&`jC>`tn=K1PWHcX2@(v|2us!fb@7CKw8C{JE z`?*~9M|e7jF$gtyv6_AvQ_C|(ay;og4cm*<7{xH7_cy8DJ}m-bXen(>BZ^!6zV1(MNo@bQfAueg*;#a(Jhkyjj!e*(+~8c`&rVI53Puo6 zG@05|_)&xa6nzU-Lk?&i+#U)|?Q;}|{ zcp@a)vw^LsX(0MzJ9eAU75Mqp%@hO&IQWuObO$gj?ykyrIntu-js(Ih+~V7F(zQjT zXR(e)rWpGv9Q*HN?e1bSJd>83%lm>t&xTyog3fyAK)XTvkHQgqJ+wOja=SydK=Cjr zzKUJJ$7Mv8mr7(Lw#~n~xmwQ1e(pP_LY#*%CMG#PmkjgC%HVr>!;>xH>QK{ym%^ZF zUkK-%l+S)u?%^JEwKB8OU_W{Ew2GxR5g#hfzcV^A`TQ=H1c3-!{*D6L1QOkmZDewa zY~|TVD0nQxo;v?z3RB!PdP^$8TrC4s)Z%<9smAN)=YFkb(BvA*{{!$HTKvAeDjO!T ze+FT+OF=}hnblVr8?;XBzC@4@bO(^jQi)a7h}04&oMk>PG~C=@Vo3y`_ri$Sm88u0 zEHm+#5^gp66oiR!9`RqdMU_jBrnAr`GLn-d#)dH#!pjTYc3i=SujDSw@eu0j0BH`8 z$&ZYfpAhCBSL6wStw7m<{RfPz(uPyrBIEmY#7B7P0b+tJR*P0-2xlR%yJU(UJSs#T9W)wixFD7rChrinxovi( zM_Ef}t|hsGO@dK;Oshs*64BJ?qNh=VX`N`$b zmbLnV%^iWf4=Sr3@JbmVHIegMhq=Bld1s% zcCTM$7$#MV*d=GHh*4{4Gd*D=Jskq9g}Ycogd6X4GPo zH40{16TnZ8`V>ihJOE%XiFI9#4`2I)xk|waT&aE61GC;93Y;KiR2*Ic%!C{eY zNGGCv?RU22xnK&n#r$N^l|WaE?hh7v`}^=7?gJ91exbrq#cTy5fRCZHU#7P7aF8>k4~wM|i&oRAXRse&eH|(N@`YafJpJ@D5OO4HRaPVPLmv>rMJ& z%D7e@rOr%t^7(vaPRt z?pOcVbD$h#N3sT7pcz*aD^k6wi|Q=$AENs8p(!n+{ltIYtF0oc;z0RQ zrhOie$_%cpkk^&^kNk2pJ+ZYPL3hWtT%52TA{)Zg-vZP}594&@ z{ah|N_EJQY{ZI^yno%_%XRpf0#T(68;9l-7+0?fOyO=sTSFI=9vqT0Tp(Xu#DL!^P zW{-3G->ei`>gsZFaL{g=xK%v1>g`cn6OS)IPHt!$1QakVGuZ6MPwt60x<_7c!Lv%P z;ecf6oMGj#AiQvt9K#!X3^kOLPJpadh@fz$;xk4_kjQdY8w4Xkic zf$%3T#GIj1k9GHMP-7}?#UDB{B_ri=4V8p7;h^-q1kW$_oBVV%2eJQ^|IXoNPD$%9 zrR<3(2NJ>b6YWW}LSaV%Nf;4;qc_Q3&l(#QK}m;L^R2U2aJ@>(I=dvk}%;4_E@2260_QEbT-c*!X{V4wM=T{>RrcToFks!g~~x*L&I&$ zT(QwD3c(X5ehVAeohR)?O}`e26A=fWSDv zJ>N*y&rbCg_xIfj?P|)5dx9{M=lF=lK1a4}(?T6HP)rLq#U^TLqs*wJkY)^rJt7Xz zPO5VBy+E?9=YYoP1Y}D1a3prZA)XJIU0^1hlG$?LoauU)GTtfGpo+pi^e`9_7457P z2@A{6eQd-vcu7x0d^Ye%R|){{1yeBJ0Ou#;@rdI&5^M_ znX!8~r@BLw*$8UuU%-oe(66YX@_nUN0}9>f5xxrZJhfG9Es`d@%D%5$DY~tc(O2^} zY@=xj?-NPU+e=nlU1E{8j4&xnMa6%}`qL z_@oL57CgXP@bHVU-fxEcQz>+?-o*gonE=4j7TLO(vlPT!Yth6|%kvx?) zA1D7f$FOYnp)0oIca)oE%pL~s!+yr%6J^@}ARrJpABJc6p~V#c&h|xU3?0aE#kjsa z@T|Tt9QJj1HP}#{62|7DZZwA<3+){4S`)s>bat6*wJSMY<#l3t)>mJN%apk2`ML(n zC=$zANd~mqfo0;4*rn2r7gLiz{-=0U>DT2z5LI?c=l3}f8Lsk*0~7rK&5LE8-JG|O z`Iz^!X7at9WJ6Oe^1--p7{EQoRJitE+2)E7mOp~;MmXv3nYyb8wJTdLFc&8fHx$Nh z#Me1hUP-9Pa3LqZ9YOAEf@rO-&xUy!C^1>hLKf9nfU3OhfC|o7PS?ov`{T z>Cr0+Yq2`3oy86GMEm%w1Qrc%c%g>v*=X<17(GH|n+PIQ_Gt_D&f}lu})+!J3_%cZLehY8mya(`v_NGU1A}=RyOnx@47NU4tsAS3iAI9FPO%pBX zwoKc$ZQDkrZD*xz+s2zVDs9`gZQDlg(@$TV?midauUHXd#+Yl#4%I%q-tS(VKTGv& z{b^!GaX^7zvn7LH{!UnL-vw{fNawC=B44CY`q)58+?kk~{#FdCSO;7@h)N;#{z#;~`KdanpsPNU)f+rnVd`$0v7Dm@NKyARbxmc>U4Q1g zNLgWhO&y;^G)lmwrt}dHPt>nG4ERgun-X2nuJItiZo_z)(6UO}U@93$?Jg9%8U#}}1 zS@xmU!(Y@m#Ar)&@Xr4(1Jcb8UrN<228_^%XoW0DBvIz3!SOZe6=P7lu#<&nTam_L zk;3R{GlHgAefHd@yv6euh>?jik}bEWZU;(1!DbAVqKJo*b!JxHkEo-DVQrnV7OfI` zfi;3cHV&_H5SVI5=7#&K)2q%^{85W@(OWUmsaz&a{a3K+dqL}3+>Dg63}W8fCMvXU zBCL-Il!pe5ZI%Ra4i%(=2H!IYfj=w|GKdAybJN;27Dq*kN%)=eYB9xnbr=%mv}4B% zZObGw!N(oZDb>kqAY^o=_{|kjxl@Fs&xc&hLWQDmFj>eCkN-inxjkAPLViwT+O8n7 zl|gQUFjpX`M?zZn;_J?O)u`aP8OYiU{CfAmZ)endG}t$;B7s3-y>%zsK#s30oS?HW zjwS2qo(8E3BJ74|*cYinx5giSFWE}pI1C!RrNj0niy7z$JyHM_!6)@ly>EiWM|Msx zPunu;d3o4h14Y&qDw62KhrW-z=(xpQ*zlCopb~1IMI%vpONB6m+wnJzY~A0TeNH2B zdEJf#bdL}8Q>s8Z%kc&larNMMz$M-B409~0!NT>P9m)NkTQGm`>yMh0fUQ|Ajjo*D z((*J}hR<>zlq)mS){RH=i_R~a=}OnM9-EBRq*3<%uAYMBb-doZYGWx|hIcJViUk^9 zC-&OLZ9BT$uDQxcjv#Z>6X|L4wHe6*%z*Xg zgN-}hPi8R8o!6Kf#eJj;81farOKgrJMh;)WhwV3!XPv8gek-4s$3U&GRsT0a7xZVi z{{zhYAJwC>muQAp@L(ieEVYS==i$0gnyD_Jxg}a_Rl+U2-~>=EV13CmR63x(p_{{X|L=@ z^7%Qwf7)Aj3J0O4@_|ZTF5im3@2AeiT3y`mIo+~x^@zN)zk2QIR)93RoSl5c)AYnc zF`vVEQK0`csc(zons!zC2*6Kfwv2^8)SiZawY*2hL?hF08~Zk|x4SVw;vaa-KWELV zSGKg-rG((nrB_C_u+DjQz-Xp~%ZmC*>9iN%1++AGb#j5M_FOopUf=eM%A2LO3q6|Y z9qaj;_SWgMIRlnX*YX#5cbbx8m0oCLc+blD9m^umx&&O^49V;+h|><=#(Mn>Qf=b9 z@{&f*V0=Gjn%@5FqEo&}ujGQ3Dj2_5d~8`@zkoJ)J}4|xXbHTyR?~y4X=(Uff}GqN(v4HkuQeT!*04s)rc$wJQ$KpJJO4}tl$=1R_>vUaT} z`jQEKY^gr7S;~kB0Mc;%RIzeGO=67?E5VTUUn&!vBOPxU3z?GI^_F!Q=2CJve&IIF zq~ond!&TWj5?EFU=b1bb526w&)`+S&(u)qE0FlCjN2GlUJ)Q`!yKz2O z)D=6ks6`sd=wEqc#j*{4?1+I~Cs9F=ctV}y5|rYY%d<`OY1`F+u82ydxcX$RD^h{9 z8@Lb}iY@YgLxUvpEr?1Pz?MqS1Ux)HM#M$l82h;o;-QUB#>+Iqhp@iz&w<359Lhx9 zvg*7NryImJTyhz!;hpu{S}-y-rX?>HwF`-fF}I?3-vZKQ<=BCVh&4Bn>@f1IzvUOu zEYsetCNZc2g~A`jAm-yZbF&}0VL~(VjFHzwwPkm}xhb>|?U5OzAA|V)GafEt zi%n`-XX%L(P!wsAPiCR5OBEaicO_IoS{dr1ZYW8<7|_?2nVR#CWtG}6(h@ffdXgZv zrzwV8!agmr{U0fjOB`=goAzV2*c=HCpbOQDSt2Ew43601)F5tOQ-&ncDvb}e%Bqq) zKBh%mRI*FNs@?fZ*4}1Ffo@y7-(|?9V`hlAu$M;PRPK$C6GNpYk=f~Dob}RVl@=$i zEjL>FA|E;xR+2eAMk9uopQkCWP3f#OWU%NW;u9a;>cymDHmXDMnlgfNR9JYM#t6D( zzbx*Wk|BMFvCoB|shK8fZm_D|7%lcH7X(U)B2iRNR?~)U$o!PbUSO$a0oAzL?^YEH zwp2Hjiz?Kdjz1^sY~7=2yo#@)+R@~;VT)DtP{&*WNeEmedy68ZVpK|ZQQcI4Z()(`Tl8j)0LzOmD zEI$FL+JT!y!g;mIH1y4drNgIQB*c1$1z<#{_5hEiG7nXxao4T}eVViFQ=7aUSzX|< ze0=@^*~hg00z59TtR9(Csm9V4i2YmL$pZa3;A&A>&RX-C%7Sh83)s^+=I$pH9>V9y~ybMX;KD-6H-kcY(_GBg+d! zNG7H3da4wd9zuLu&r7EVo2~f>AJ11hmd3F2ecWZN zfB(5Ayo+izz%BJK2FzEr`Eom-48yaq{mT2HSWmfO{)VPu?UK1}H$>V^4-cK~ye04I zx50aYp=0>0j`Uw}Rbg${*eN*qeOzym-{(9_7FIBaSBjRus(I&0#W%@-k^3~3oHOd8 zd#zauH%;3V&u3WWik$NClq_>TpjVKimy}S3Y~nymd$>e#3~CC6WIMw;$flYJDoaH$SQz>|S^m<-&r2Nh^S93-N9tokDHg=rz!X16(Wkizm*`k#tQvnm5ye zB8NjT(*hd5AbgEl9%WnRLmIo5eKPouzPS!Z?WPY+av}l8bItlub+PyI>~c2;S$f7> zAur9Sp7Zo|1;S)K{&tOu>eg_YldcbAM6bk%Hct56xPp{U?;x8*#8D%8e8k-yGe%;1;^9m#r z>^}4(vxEFypMm3PRxbA{*J*=0-a`f8VbGscqMKH#zsj`UNgMuas25A6tY zWL4k#%O2t`B zK3u2WdPrM4+~<3^$D8KcWu9_FkpeE@_oxMP#W-VLbCO+6f%V456XN5~8d7QhdE^Ee z>!i$Nh<~<;p5t&}tt!{RN?ehKub0@()Ae`kswDfdFvF18o*(52WJm6>r`0BDilAvY zvt``DX_(FMTqx)MaL9)xM@Qhk0}OJPy(M36Re(sc)>)kdbe9lfn_LRrBEUD7{dEz(JjA(DL^I69@ojEW#o))@K z7Q8=8M<*fGF^smuH}QNj5_^1fzX|nN9R)oHIw^Phx=cV1$Gi1AFlBl1SzaLFDK0pz zS5snu*134c!#FT9Lyo?SqMIOri-|62i5JQ})4=DHDr|QR+x|?VrR}F+M-77qnGwr) ztkz3hYu8p#p5%j_1;D#i)=e<#rYz)W-W#C^Aa-t0cHD-0cMZjZ?LVv(fs>X@nasol zR>$l1_v;cKE(A%vV_GaEw`dOYc2jvj#=M_7FSaTLlkrTbhMOK1Sk2hxHFSs1%=_ud zz>&KuZ5-M&X6&VkX0V0@n6gV-t${bfEgKp!XIb*=$v>IngVVUyt*Wl)Bu;uR z3tPh_KYN09H_bny?-vaWl8LJ3XQ2z2lN8ok`gj#k%^6Sfnu@DgjPdbd|9BLlj9HuESY8i0H0;vbkptnDvG+IjjR=f5Q7P*txQy*B{@`qpAvoLqV)Y(z1Tdewsj(Z^)*a$!aOFL|P=wGD@Wi z!+~(1GB8%=*O_SYww}6nzck28?S(bF#Ql8#i3bTG$<)qvci%Kc89`s~KY}dJwrA}I z8^t!_G12@xn^67;Ds2~&l!H$s@L(bX2{arD9bOBm*xT$ff%>qSs@?l;1WvrSYF>Tk z(?xYzX7g-!R8A8*ba3hUWi=RbbC5t*t$v$^tBzmWM>xf&CifSAPQ`T_WJy61HNFT{ zQD+z@r@c@T7&|bqMQ_>Sx3eAXhw;h3Z^T+SV_7eLU6nz#@XVn5jqtVF;!gjp9nO?D z#7(pJg%L1`D5`HWYRI))&f4xtu>9Y|nWos)ciIbkGr8xcRDE{kFphcP1~FdL(?0q0 z_v5_ExEgqDW>~z%NJbmnB8?h|=oh90NYbbOvL}I%C;<+>jA6j%0-0p+j#UdLWucwA z&XAVe8>)!n>gljxUj3WcSopxEQ@4(zz~8}nR%+P6XNqqOuLqOw{nTD1VB9o>>~!k)hdEX%Fw^kR6^0zY0LYoE z%lo=VG3Gc*q(iwR>+zYcPvRDcZjP)?;YD*h?2u5SQY9ye@7M10WD#seBU0{fW?B?a zSeAzn7jtu7(v^Sd8Y)ee>#q%9WV-3IqwPZ@?=D6sTj`eXyMJpiGpqO}ZklfKULE^5 z@aBwcf;HyX$Sqh0y3YBl?ECObXW82-gMMIc?&R2S`}I)W`&=WI@AkI8A?NRc3O3Q zh{dGM|F8HCY*M~@Vn>d<^=}C$qoV~gXjr?#h+kaJer57bpE!NbQR?C&FGnUX0_vZ$ z9je~#{BcJ8Q~Qg_2_|T}2k=80n}3jQCZ0J{Vw^7Sqor5+nF_-Wyd^PD<}W*)Mh0E% z>N#I1-3bos@M;sXAKt~ivtK$poNYgd3JaVbnX6cgTIHBW|-m+Pb>3`&TvatWZJkM+0$vE0pFsvkhP9B3Q}#2nOe{&eFBs zUZGE&+`ZJ5 z+85I!b+9?xnj!RKH#cbS=_WnWo=TrsxukOU;RT{EPjp#MPg^yyfSlImSp3itZuc6i zt0!vyvYh; zHt(yjq>5TbP=v zOHmR$TYSZxNs?v;YMaFXC*-srd4kW`uF4l9P<%o$P`#_QHw_t@%u1_tF$wC`KcsaD7gwH&QmFN>ut3G2;5x@)|(kkA`bq;%SALR3g1@p5XD_&%i zlXblu7{tXOZlP>(;X1mysOCq&3}Lb%K;ei*yv>L`YtMzcaV|uzf>kMj)7hj)&>hrN z50=A5qdA4C^!2UotI{)UAjs!v;6HJN=2|IyhA&hC)~)MTOh%OAvs;?nU(( z6ZoCf2jnLi-_OGnjZIUsW+(`5xrlgQ2fP;_e3~w=$yG`aEdBk9aeGxZdy+5UzO+P1 z6P;jDqd7Vw&u6~kGTLS)?)YRhoH*|il?-P8DipYFQa-l9Xti7#o@9upgamw40Gj*R8x+x^jLrtj`gI-Gs-(IoAGIkO zjcUFv=OxmuMOr(Szr1~y;n6zGu>%5-GpJBjhYiLXnPH> z+OVSFql7C2(T2xuD?ZTi?Uk;Zy{O%h=HuQ7WIuO$ow{CsEOXid6cVtw^^C1!B6Gx} zbl1uz>?$2KI4ZlN@}SwG#D7m<$eGz18&^7;{m=PzOtMdIAOc!8r=oqQ>A^dt$;-wj z0@i=gLsq`HhwL&?(CIsUH1gu9w$;^-#8K5Rn-$UbsB!+Q5n-e?C%C$h50ks3OwlET z_6aVq**b{PiWL_;tnE3_18pL?a<6%EGl$%UxQH?_N)%;TJaeN-ss=S0mi zxe#O&%at@17M5X`>jKsctbBXkD;X|~CFrLm0I$u}%@NVxJ7A*_e3grSeUF)1q@mr_ zN7Ll8haoAfM({`d{|n zX;T^4p}`uc9ggrdg&DMq8G_+IY9!tmr3XV-VzOp7Xxv0tPm4EPlZ%H)E7C1yTEdMv zb+^fVc{6P>Waz(5ld*T}$$Eh`NU|-%pX?sqh$-wU1ZLw@C(a<4>9Vp>Nfup_Q~{zA zosL3G0M~Q?RLC8gi%(Z)-oZ{G;q@UtHt(9Hl2o!Ib_!rMGy6eE5Bxt;QOI^a5!V>VH7EgC$k!UPg6C_-2RO)p-^K^2&L5PcFa((a> z!=z3a4_bfFhp4N<%p=1>05dnn1V~2|Fhy8Gv&rAWXC+cxCZ*Ndn}?(xZG;6&0i??t zk@bPE-J_8a@vbBdnDC3O-nL7~E(bmsIwhCdnbYgm!bxk7Unv$)Uq_$GR8eArXq#@&R z52qthXlvGq2eZ`I7l$Lk80+gZ?d*8{{J;JK$e~;)*kQVE_k)D&m|dDdY(*?tMH5?z)+Zu- z*&6`SH&l%b%SJWCW-b3;z(BSf&4JsCkX)Ls(8C|IrX?d{TI2N?qj;gkQc(L3pEz& zEi_^bmN{v+a~ubQt-iJq*ib5yUO%b+5aY}dgkI{p3&fVvb(~QoN!&(typ_9T&30l_ z_^p)JqIGND!u(IYo1`E{Wz%L+uDm5hXF?3IKkuK6R_V}4_@de8yN-E9CO_eytaoCC< zWeoadC&5pSG+6y}hV7mnv@Mb4&&HS9)){iM%a&bPQlVw0!-I~}#`3Avh8vOnv+}!u zoU(0mYsI=Ma&gZKJ%ylVW;x2aD;5)pOg({GwF-O_fPfjX)n&$tXo?yhhSDGj$|o-D z8_pFr{^El?GmE z^eGKpdwLQ0r?jiqmK!bqSB9IA(K;Xb>`2_I2`)`PLSv*mFy>f>HOiex9Acj<8kCPD zJ2l5gcT>HU!A4(cRWV~ai7aQ>*u<9Q6OAvZ(`iAIxHHS~{%)FUcROZsOYtlSP64jc z9yeF$h!>&56FryrEaI~iA;`cBYmXk`4kC8itg#pwyWtjbdA>$xp?9Uf^f{ve?i~!( zLl0z@i~lVb**6k4!2VvCzK7(a0HbRf-{6VK2C|-tITdBRsBSy4vhEsTx0rqYx$U|X zmqpRPEqZgFf5~N0FtBKZjwp9yH>#1)o0(*dl?Ja&IoE#gyv?|GrJ(hCbpxGI7+Lhp z)cx-Oj_|?+JKB)yGA7FFts7c&{m#xYuV%x1k+TzmNYW&Mw?;0;BV%MYB{v+b8 z&o(coWa3r`ISwz$`+U@WV772k3kDma#Z014C5Q~2y00yAE3Q+LqR%v9c<$=CgGG*A z7fuf?^mRoiC}U$wr`@$;7Qd#$7&c;>ldt}`3`m9s0al<%o3oEW`IAb}?q0jI(rEHv z-DY9PxXes~OyF4l#Ny8fQ}(`NJpW%g>s-P@)aWa!T(qbvp_D&D7>C)_($jgrj%Jt@ z99I&<4I|5uy?j*7TS7PZCU?}BFFZqVTS(;vqAiSLN9?An0;8)l;ZEX#HOWZ z{Lz;*lFSta#kfTW>bHko)cy43fHO?GJcCM_Pm_H8T*kbxznkq|ab%QVzB+2rcU;;HyUCNo z#w{3;OOO!fKP*wUhyh4hh=?q*ceIjYuPM#x%?$vZS^8Ng4uMM(kb0H9oUu(R$%ykj z;KJf8$i%-YX45Qd=kd$8<(&oL#V9kFilP1Zvxok5k|BEQ-B3dmm>$wHjke=^F1+IG znjIqo%RX!?=on3LLRbjstrpxXjjkfK=k<&#YbDo1X4hI)Zln*FVHV$b-|Y4AKRkSs z<41JN=bD6)CU1}A^x%Hi4w}qiXP$E1{>IitO{qps?KjtCOF^f5oPRqo0dp*y;@BWo zZ~X!Mn~Va@41w0s{AjScAzs2xuh4~JSI9o?`#&)}-Q52P;d3Y(90(3u^jTOeP5V`gH%} z9KL>1=qUR$3O z{O&g|C=Yr4f80JE)bb9{a-9Czzb4No$J~=0nwoL>cK$rdL46a4sGg7h>$QLE?fJfb zO0)Nl;LeJ84@}Q!mg*h$6`&C~Yd2VVL$%?>8N|ZDex|%tf7*ZDK)ImD<_^ zJ7ES+6{Mw|0DH(EQld3XU73ASDk@sVEOhK)Iz^13SGA}_l$_AQpt!_#OA>HdoucuW zNO5n@n?!i{Sx7#OfTORaXVK`}N_Fwf$3N-DGhesR(I!v8cE4^(ZoR(mdLV*V+2s6r zlH$g9&<}w(3+oDJ8|fxCk>kE;j8BfdZ~<5~s2_6Uf=9FPGd>g)JD#VoImk}7`t61d zEiS0gpy4D+2U}CpvTsh|voABUu#` zoG{uaEZ+~eXNXT}!{0CxqHkzR)L-K0wiE(+Dc)gRSXvw7^^`|+nBmetB{>=u%Oy@1|ZLPv>!``ozST|l$e?C zxBv|sgnt-b_AQuT*FCUMaR9}db|i%dTjx1M2IZmRI|HM#gI8>mz#M<|VHK}t07DEh z`7^b$m&~XVp9TJh4IT9op<>yQE$)yR2}eEY9~Eu;%ZN$fUwEmCVl0W15v33!T<+nz z(3w&h5b5%Z+@&D|K55_w9Yu<1gDFMPtS(Ah!yrC3UK0FFfD|V|q?Mi7d-?4{it!|6JQZvD zi;|6F4xBqc={Sppwk4u!Dfq(*LM0-rARk8|=MCzhUpPM{ueja0JwT}5J|B^r&)QE1 z63w9NrA@Qa=+aD5eCrh=Xjnj)iCLc8N-?+LmK}JhJI3SqzguOj(G*Y|X5mD98KU^9 zF%`6RWCF$Rgt`U|5$9d>{n zZh8m};(NyH_07NcniIpZtyW{hP396@uUV)%H03p$Cz9p;>wtXUFXo&W4t>p?=hS(+ zZZ?%RT4OJYCsS)&64F1TGkcbT`{4ot%{KHG63u2WMo{}RpAmR?mFf?#JcZ~ajWW68 z*G^9Bfq~FLt|ZqVJ6DJ(0!yj>9ZrPsiRcBD!*+)e$R92KZp~T?GSfokumorB88pgs z2NpVs3+$Ug9gd@>K0oSV`s1j|_^TFA5^Zltir_E=2dk?{Y&|BFSXV#?K7gw!OoUl- zB7XtEo>?xSLsV_oAe(Mb1jlIV3uU00_{Vwl2M}Q7KY%szB^Wp?N;ygh7Ru%kwGva8 zk1!yjP(!YZc&d7ku5xb$2JwoA#Wel_gr&O$49DQ|R};PIRh0@l-1z3@K%}526qf0k zSvLSZrfu&)UKh&O&zYYtV7N969FD3GJJ6Wde6Ck^>W*8+PL*jvu zj@rTjIZ-Gl_u?_m-66tkqe4q9#8I&|?B>5oFRdAV&yxPhX0DA@ixQ^?-(HPyL~+ms z3c-DjIO%OOYVU3O$%oTtJ~HC^SG~PveTg!#4O!3 z!P1jhp21Cc^q5M@tr$mYV~_~j6fsGnU;sTp6$7>EZleB7r$CQ3Sn6cWWD7{#OWys+ z#$bp6BnbD>NYckFm!wB6y6@l2P7-S~khI98W_<6iXN?;; z^`!P6Cq1f9m|MIDdm09ez{8S^&`D6^-?8I@aIUDLOFNv?%X5U#Iqf^@%rebv>dRy{RUeZ{RQM77x%^40Jr#%vehmN3(Wmpd5C(nH* zMriWNfeM;8@Rwrw3hJ*2@3?DexaRkhFcsN|rOyj>7^fAuX5tz`qUkfWfT8`U?B9Tq zdyZ3MjN=uZsp}0^c(ThBQKvY{AKPse$b;jYZd$hjxC#K%|EWY$!w(4bkv}dB#EG!6X=VO;?Gwc+U}~1DkhMeP0o=O z>ol}hch?Rg{(#9sHTjBT>;{Cjoe0%(frW=sw2fcX+zt??X2z*DN1ZrGNSqfqX~X!k zuOa11Lf%=Cua%*bc+lB^;|g_Az7nk^BtqgI->(M2r^n-NEqvTFtx5R}XtO|Ie6EQ@ zUQEud*NQU^_}e`#Y*5~5db=R%_y_>Xn@&7T!8^)lYR^72$b80$oqExFV zCy+Pw*rjBN(KPk$_A8CkFI%<7-N<6ce{l;(_!Nif%MyQhS=soit-|*i9@|&8^R09_ zG%I$sQLtIM2pGI^<>|697C>yx=RLqiYQe#PZrzzI*=Vo5*V>6}7$^a}WjA zT|IeRiMIAg`#mKS@MC7Z zjNaA71*5ONPOMi4#Q2ETg_r&lK4Cxgcrk1z-dp`&+wgQpoQIT(;E*U5PupZV2Z1k6 zk%5ICYwgnty}GKb#7JG2`3q21*GaGpG1N;l3g9>ypJa9Gg#u6>Go5YC3ZbFC?^KBp zdi@}(P_ZnignD_}Mk%bY;rePkHShKqr|PVr`YIzRFSA0Uj{7O2%o4gjz>u8u!89?3 zY&KLFO*Zk*QygoEC1WgguDd~uL{w2t5r77&?d1x*T|$`uDiWuBhewX!pI*c!YBhFX zMXv#q8x>HrB_+L(5~rAyl;Ra-GX3Fb*Pzl@SZDMspRhZ2z6^Jw-FlOtSe4ea!sFh_ z0J`kcs=n0k@m%Fu#C+g>bd#j@*z`5oE?#zW;s(*S10G5rpIeQ6!{2vvh5}lbc=01A zl;J&VC*MSCT}?aT)H+jr99xOI*21uYnqiW;k-6UhqF_*s74G%ewC=R}88JvtXH)X4 zWpC7tVni5`@W|u71(@UglYr~nW_s70nJpRr8f4ceu)9D3V@F1OTY<=)o21(qJu)F3 zIrGTBakrIh#c_UT24ng0ry;OGP5tzZ4G`+m|F5mlRiG=2=ySk3k*;u7gs-iOjl!!V zTKb#(+Epzgh0)1^75E1UoM!lN{MLdaY1f&26blmNxHZ zu0$Lk_Bu+@_Jc+5zNgzjIAqqhpn6FB?;KzsuR`SOlnCO z@#XF2b%XeN)V(DpZOC3#0HDgZDt}aKPL#x{TOHU+SrnNYr(+hTvwCdUmER?j03yh+KMw9NUm{IowSGX~5Hx>-;}i!;mUZmLD5Kd+*J-ToP&^<@I?wkxvP z$`q5QPx3mGhQ{`+$Yp{1YpK2_(AwFRLs4uyL#=}uSj7hVJz+rGoxE#vyB^+(UKo=q zt{;RF--W(?QGAO?iNgA8_KNXW1O@xa!M}}%11GhnM8uKm-Tk#A57xg#m9q^#hz-&n zrc6}7@+ToR%ob-$*-;@_|MbW%m)dPqNp+Dsmva|l+vCVbxSn7eiCF(Q9>>d1r^cHw zey#fg>>8Fx!B6UkqR_d4UZu&gptqAn(JXR6Ln3l~HA*!-*btBDiol{fo$h_C1!`XA zD^XHAP583@@~xz<4WB#Q_E^+FxpBEqkk^NLmgRE1&cFFtWgBol z55rnr*D)rZ{&!cQaQzjtARK?xYvNm2yqLk^U#kA%KfGUalgG;}DZh>f7fU=3y}|5= zWcb}8!|VZPXuo+$vPGD`vq_e%JR2_t1MD8eXH9C9^nDtC&@(^kzMijVc;*n7!UXSK zCR$s4cJCcf^xiu-R1$ZMF?S&YW&>JAcQiB{CVvX##mqZvr}naLCN&cVZJjrFDcYW; z*L!&0slABzH@R($L#00v-C)9+@5Ya)C&@IF^%rLOzkQBgA2JJsfbxnp?kxQ-Oawk3 z$~B%elCU#+!)a71mhSjH>8)V=|1MgS{MTB2Equ`kZC{7cZgLv^ol!=4d0p2#G;`YY z^YiZ)n93xA`~R@0a{rGORW>e;|J`pm{$GK7JN$ov{1>R+e5Xvn{xFbqfNAR9O%UDI z_6uFm|BB=z$dpmamDisgX4xpx@0&YyjirzFRZ{+M!vX4de6eic_u%PX_CqK8=N|Lx zZZ~gG@IF3|clYb*@M%==GMTVT<>u+oufeKLKBt(lDqCJq;MWsFB=Zg@v`!uBkkNi3 z(?N~;;lFQV_VsBQ?u**PS3Nvs^Yq%P7zqSRYyW)Vjh?e&Co_sQ7!76p*&U%Y2zAKg zV({VEhj`S=^#-pvmxtV4le$B_&1EqExue!a$$Lt46@Sm-`g3uY^j5?8`*FXznCC}V zzliLX*(dMH5{!5tJ=LZCXzC$}W9l<_GiH#r2DuefxdGNwviw)S-YlHG%V4J|u2O@u zi`37W6*$2Kpm1;thSwSh+Xm+7m2Bjvu9V(}{RgLiNNXboK}}p%ScR<8Agau}ee*zU z0bUQUIYv&UhyD+@pzvgLOgt25abJJN@k(DMlQ&IKLz4{>Z44{e80TAOkYfu1P>2Gm z)KSXcHuE#JtohHo^%kRF%wI4;PWnEN13vEx{(A-(+j^y|038l1hL!bP}EJO^jkp~h$Ja!tNj zT+Srpj#5L=EQ=K3%3WW7usC@0NsuM0V`tn%)G=WYZF3i~It7eJqi5o!!ba6bwSZA| zD*#i5>aFRyrsnrjBhJduAsP+aZbw#z8G6)kO-GhKCrmW)NG6wwoX{17SxPr(gr{R0 z-m`0AJ*KKJch$l!l)S*<5oEZU!z{&Uk#UdP}?f(|0rwJzMDpS*}`p=jQ$Jn1@ zSn70jhj}7o1-8N%)yU^lo{Tnvv^)iLs_mIVaMlMYR&YPFKY;qGKp`ZkxC0kmQOE<7 zY<_62Om@)VfW5Fg(3PA->pN_aXX~z|K5C1~@gGD$o(Y==#d1&^QLKNm5U-FoG_ul~ zl040n$`+ChNi}&SPZDCfiZOS~S`o1XqH#*NYDkjQI~FV=In)fMj*zXZxZ{bkV`2Q2 zG-hcF_XKb0){dqiWeYo#H$HW97??E84Hrd+wJ@*1tMFdhURymwz;arrB<>SLEy{9M$O{6zE&`7?Ug5#MXxobr<9t(6 zK$et=5pDY|hNPVDzU&;vrQZ7!kipxX_z6Jkzg7%!gmDl^Ib%i%sT5}+@9qKl`u4%J zhJ08xHD36~FRE7JF8LS|k2t@~EE}h=W;4Fu-_AqaZSTgMXvZ-*p_kI{4twi)lW`|D z*w+!a9)`e2yGnJ(%}RZrjxw513^vNILflr}f!%)@F^_8Yt8(3Y=Q7+{C3;6+)q(g4 zrX))AQTkC?ml<`e&Z8JgC4T-amFaAiySvpiCNSS zWoLnQ`b)n8KY|>^n#PU;dvbqeYTKfdR9X}s(VgHDUo+u=sM#*95zWlWt(pU21m2kE zz<@ebwO>98$gU67eZ<;%WdEgCRqHs$s+*>r*XVyq|xEIuBu_cc{1m4e(}JBx7}l|$0xBzK?@lEB>uEaGRodC?v9x<8+1#u zYe~Y+sdjChUg=rOQ$lcRK#bv+O2YDqR(~)55=SVh#f2+c@=Gwc`P;IDDU8{JNBrSP zWeFy|@}BFi)T(?)YZ-whvC6aGWNuUKjGj+@!QQnc7EWtJ!^Ex_VADPSZoSU9eUlm# ziL6ctqeTt9K(WW86h62IT__~QKb65yz5~_{q1G#guxy_jNkwcv`%kV=Grqpm1iOTt zI6`@u%}d$BA*+deg|`|{r>9Ce&3z!$c+47p`V5DaJj;W0LNrTaB(yY99!Tw=W@7Uh zri>Cf#XJo7zxT5EDPNkKE3uP3c8dufvrwrV7p43k+l6_BP$J#d$(WQp10^iP&Z63> zby9{=N##z{q z@`5UIBkd=HR6PU*m&kSp869k>9q(W5+9AG@9*8dORqrO~#tjxNw&6UFzc zO)j1gqSnR!HC^cp=*#xKVgtXm-toXAddBivu>aR4Pf)_6F4meT>s6G zaKA*j1)NMsUh1*ko}`ixOy|c-rz-+|UX%l`Hmw-XtNfE&LCb2qM%P1}Pq<09ro>V~ zbTjA;Q%J`3v3%GarIH~D8>=H%135#lJ}3ieOb3J`d^)c`KhaoC%c9eo7(%e#dW;XWP{wkbGXzsj z7FWAodTtZ9WH$wZ^eqzb-|x+JSEN8jWD^AwJNnGG9u_=rQrqecJR6TG1NuJabcJf0 zHZQ7j9=UG+4Wo5Xfcbgn$WoCCv8^tDpwWgMfMXAu{Zuy7eQ0K0URqof-<2@Cl+*p<$Z+-a7~v% zu?Gs@-r%BXOu8pz40(|CW>Z6eV8 zzZ(LZXMMHJa66v0@nI*+#z_%Sbul9|mJLLE)Ys;hjP?xzl)iHz{a=)QQ;=rc(ruS* z+qP}nw(YO%>auO4%eHOX?k?M|+ZPXK$A2F7eOM9eY0ZcnG3UyWnRASqciHHx5&Qe* zQVWfr#xfJ1&aWZCo)2NlQsTf1NjTsb)?$MF2TG9CM_@@UMr~ZekCk}2>rXHctjiwl zd%u{5K_oyWqyh*s3ax1tR9i$BAn|2sgyd@WCLwSLq860ogCbf|!I8cAeEZf`o=_G|OfNngFLE zJQ5->BF9%X2&W7(P)Zcsx;IF!o!Kd+t_`jM1L=ti3t(9eBA^B{6bOaDWBAi8hsuW$ z>t6LECoJ);Ba2lGc+sMQ=6BfYcUPdRWW1nMd>~E|Br#$!zKrj!%KD!lwXf5ZGEcZK z8Id}TMYDWZyYhu{EjSIw^6R3)AeTEn>Y4qrcQ*H07>P~NW<48TuhJmG4+J$#33 zocMs>vKKk_8D7|T&<;MCe0eH9s5DKmE*uI~%l6rzC}#FuXpz#F?%01BQ!DqTpTfE% z(5Un4P-PT%-!%b|b3?AQ*2 zc?;m2$at7YtBBLb1#mn@K}rtNRgb2X=Od-;yo-aV9AD9Jdx3V5@Ria;xX^Eicv5PS zskJpowaW31`4w_NIXidhVkA1e2f>d!KKKgzQZjG_M9Ojs;(&H}D_bMZs-wIHxyU)K z(o!@_>ZnNW6$LfL<*`Ts>zRjg*z)J+JNCoNTJaSe`1FBe zNKHJH6?Tbl5?DsPnZbpLg4+Ya?Sehm-S!ywWv7aW7bY3s^ck6gyxhZU2BllfG#cra z`@R;qt5Bq6fnmAGQ84*-iugcc4J6q&Smx`Dl&L*IK=Lx2J&V&1``C~vRMcAHMR@1C z6M04D7j6GgA!ei5a4Obk8*jk!MTSt#H7#CAZIHCOq*Bgc7A{4kjsxekm*`HBH%<^k zRH@D1;0?Hhd;{f?xz1BLi3N>MZV<)GwXs{WpMYt*NCdI<(;fOk!4@K->B$q;Bhod? zAt?l!nPs(-WyTxo0mC`(QcHxf|7ksnej3Z3FZCI5E1?S(^}&we)2UEUwLI>RSU(P+ z`R+9=MwSh)k^LDzhDbCJQ=wi=RI~UtTtnlR)-Y0#{fL#`tOrISOss*eH?)PtTcCED z2qdv<^F!Z^fT!~kbZEN$B^p3WG^NB4K+rBK8?v|`tVK$x8$8;UtbH`V@VLZi;aSbE z>D@&hnFgF zQn;c51P**j6{3!`{1X72Py#h;LrV96KF}OYi2=>L#@6&UJmuiT1VA4yJ(q2=FHjNv zI{f=qu|6({it8bz4hf1fIDp>f0=L}G)y!WD3!leLH5{y}3?pbut6I){L!Kxtf$*J6 z2geh=$dix75d??#d{<`-joVx#rpG(%nQkDzc6O_X%?ps9A5GFRgcIqdr^DiND z@Xh$-;cuR|ALdAhyGo9JK3u9WMCo@l-p`16W<^ZC zDn8+&Y?h7zRF~M`@#AX3oOMY*dba?DG%CZ_f0g~>$E>{K*XO<^J%J|R*(!h=uC-?<}sYHFlh@Mp4pe|fEYc;8o;z>v6xT+^gfz-5q zM7|EVVAufHGO8~s;~UdNxKLOW5Y@f$2YwQB1U472?+( zMN~He_9W?C27g2;35d3ZO)kPlZ}JlaN>w#~)-%z!X+65)I|KgOCl=x_Ivkk z1i7@0R%Wlx*m-SeEKZ0t#b8}_@`fiRs8Ha~!!ccz1tVf|>5S$O`DEhJ416c*74iN= zzz$4D0Yh`(7`(Aou2^PIkJHmfEl^NQ$L+Yo6((M5#!y+NhMcX)>=dg+E)?@SR2%)NXLSL?THK@%U$#@gkft@a(@cI(%ZC3-+j zwwD)oxiHr8R#L`K-~mb%mCwKl>8M!!qYGEM5!V3D75pAzk%YXtDSw(6vhS;|Mx!oj zL}2mvx*8}@URevm1V<-YXHgzS-RL`lfez4gXDpt-PDp%{klXw(%tMy2<4^RSWYwTPf8YYEa zb+w-1){u}<@vxJ(#QK{l=JU~8Z&+j?Z%m!MR&khj2G;^vcAmAtg%2W*YNbJNpRNDu zp0UoTi0IjJ zSiwc8nYV^0)9{vsR~=2E`|ktGC#<}qwV76UL*&t!t>Ou_N^5Th23X6k1xjl|D(Mo; z^d~o4_4UpdYOW+-oHwY5yy4Z3e&lCQb(S4awBA)HOoZ3bp6!R=A*QV!(%fZ-s*WQ5 zGW+Y$CnBoug8?GHU%G~&JIEEnR@`)OnPw-R{tUCtMimZ4Ttv4JHf?F@#vVX0=VQl; z@?N}a9EXuq!58Ygig_K>*Ss8VA1fi=UBBMh)liVR%w+CY3t*jxL(CVRM;Qv|5K)KP zuU;ZuyhWVM3x|@n+dL8z9*hZ4GA3DFjS5&NRLs=T=av=4eT!wqhdp+`Y=sZPfP@Fpe5-=28>uDopD zg$a#z2N1d`&T=-m&7b`h8P+gnBD&f7ofgkYu7-PPo-!n-EQEJo7i~4$fjfa;l76vC zAtRYT<9(k}{_rOwr&eye6Jt+lIK3&%PJK1oYGc{vxm$lS<+gL8ztG)wPYRD2F0m7h zJM>D_xu@-xK7@oPctvEkpfWOyYKVg>-!uPD6xjRLS!_vd@1^HuRvoM7-JDbUE*NXC z^}>tU*{hy2c^4Eb0w|Qf#6AYqhh9kAd*1cy+DEog&k>E^;dkWjjD@cPfyj$n{3!>6 z<^D3G;>6NkdYk-0mfVLgI|y9LNxMGq5(QuGVIyQ`_ucSYlgbqAFYUbF&RHwF-wy&J zgf=<mA+}az2rU73dSOQ>DAZMAmRze7Oom>_8@$bLB zRg&mb6WS-a7*j9JbP-|=Pca=I$HELlpArn2?cd{n-goyo+CL6!Z~pZ5w)=m??RL0t z(D=V~ecUpf8BTS<-w;SoQCLUF21+~xo98ifCpkJ(M!oMGGV+cjg^^4;PG z2F48~Z66MlKqB+QRU&XsG!N!r3<#i>;x3yQll0z)MtuGb1hXcmK7qt$3{LH_&Li}g zg37@<^=x!hng~#quF>2`2gS-}`u>FU;><|?LvqrOxbqV~kado4Ct-EK51A>=Kh|`` zgjWtj>x$>rQuGkIwVtC;Zbb=vE3qHaIS0rmqV`2?vR78juj!`NXpVsvADpF!wsRqxh zPdTNxY9}smBre@aN%9FR!x9;ipKX_g6=BnDKvOa2mSsLcH*B$3xJ*Og3yk@xVehA~ z-(Z@ajUdl_Vy7?en8vP>RYi%}tZa0@3z}lPd{>Inm_`C#2D;chx&eE<Bn`CcA;VH8z(2O3|kqmWN8OAk0yq-7W3$_ z%doNwf0uX>Mty&Lt4VvfY1we@JT+ZqONl%5n6lSA}E#Tc1D&u+?1_!8bwW@C7YwXyBgFC)a^I+6&= z1hloL>USa(f85bt=>m+W74=Jjk+7)AuSiJ^Oq{A0KokQ9&`ND#VGSGeYPeFP zlE}_86aAAZ)VL1wT|z`%v?+OM0+i!`9hc8GlhbuHTiH=wwAI&mPA?CL$HS5h!4Bxm z;3yo_QTc$^iwZkUU)Iwp$!(6W*3HScimNi4rkPC%90#vaa@Hm@a-% zlrI2IaFr>x!gCf(6)JWuQWi5x{PQD^Z|Je+|#t#@}OA@1I!fta4|so1bWoGx)2liRxv}Al=W~*oR}j-TC|9GP5k39`~4lfD0eUL0U-<|1L^2nc?ARN%0%Yb<60r$8 zpQNi6hxBHL@~Gn8(;bF_~q!k)AY& z?K#X8;wp!-PJI}K5epx4*<|d@OOs_r7Juj-Xw!?$RLAoP@#@HR8HD2s`z@t5X>+)1 z_5TVocA(U6E9xQYO7RRU^HLy<>XLj+j-j=GKFvQ@&ECbZC6S9@1ws$O^h7XqFxb`G zL~d{EN3S1|)%VMYbCe+nLcr?&XtOJfwoH>oLFOoX3tD4QZZ2SLz-= zKmx=Kg9HRw%+mb66T)LaTU9Omg6Js9^Q;ytOfu3t4;XJ`?cm8bpHFE_;qtVR%*nF% zq6%*;MZ?U4#k

LF^ric6;Q`%HkiU%Y6Z7L(EXdPIgK=DHdU)sbkkX9|L`9OCs65 zn}bSXD$T&2;@&Qzb2?reRazQK!RLt!68skZ0f6 z8}M5^UBmb0l9M$e=SgNAK@@gY)zgTwXa~#fwFE6&m~s8N+(7R+GimLyPeO zP@+VQC|h=}vdUX6RnNe|W~BIH{WKogx&j6WfxO6cfN zN81+X_M**W)B`+ZwS3ByMl9PW_5^vKr02(wqhdtb9m0=`ygilKUWRJFzHkY(cE&?J zNLoQp+U};vqrpS7-6hu;K#{&zSVY*?%z!Ftv67poq?&>C!^7r5>GMBzfXuMqFaY zkvcke+++PV@zg(77Ct?fDdU*hhEv2BXJa}(HH3*XKZX|WV<|?RO*BFUGvmylnxUkMcB+PO1p7|sYeVC8qmhXfKLHl8g$aHVDk^O!WredU$O zF8i~P4Wl847>EMda%S6j7l{DZT)8>Mrp1}#;G|dN#wYxCwljqT5Qwi>A>)~8e{19^ z*bwFY0r&k@2Urg&Y@+LTD2l;`wWam82zpT;iXt6UbxeO$C9BidTPg4UN;>kam=v7^Yib>vD zq-4SV^&=c|Qi!xYVEzMNWAv&i#mK;o#!|sjPSK28wfC%sI>8<(yk77I8@+o#RbxfN zo!K?64H*=zeB;b(qUEEAAB=S*3yyCu8J!6diGQ@cz=KUdXz$)FaSgEHDfOJems!SX zDXYBwd9Sf7=qCA|MywABE1LTSXM5Q6Gy_iHa=|BP6j!@ms139sqY`nd++&a`bxKDs zS^8I0y)sh1OJ}p$1TE(T9k=@Af@uq)>&F+0J*q%p7(Ds=gXNBO;}H<9*Sg&8!tLbf zoK4% zr*El^yRiIEy$91~f7j{zf$;1q#C zeat}2m#&HK*RiH!jXYU`_3%9d`qDcO4?2N#qoPFP@)lnej(Wa4cq4FN9_c)MY$EOL zrAw1cCtK4}!dbo_-xgH-(?>vA%4IY?YuyYabFd9z)9u#V^4_a&p(GPnz_paR?eQCj z(tE{N5`(63`uMb!0Uvc$10qykz6h%(tFhnACWlH5Lq7>4U@AS1lg;I&ekX^JN}DIk zGAm0{OL;5HRY+bJZ&1bLxFJ!Ik&#xk%jlik#63}5Jmd`?yPKpcwb3|>wA^Eo@;fNe z*eSFhBlJDRw*Qw%v;Qkr%)$Dfq*yOeF6wWjL$ALgaox2d$^DXpAgimR!afb{)?Khm zYZQx$y-s>}O-E7}0zxgA@8y|U{PgW*wqvKh*}E$YO-t2R!RbFuQR_%ptDo8oT`Na2 zZfd!?Jp8_h!W7nl?)?$O)NFTFt{28?R-4syJ_K2EO0;mFz!Ws$D4^F2APkLyoK ztYTmiR}g*dwUqj%I+}aIRfS_8F{9C>x1^EwsId+RAyjioo3?aOXleC7~vfNRi?a!pI2X;*v$wpd#qwWTFF6z$_`L^a(9_;l(`eb9xTU1O^Q8 z1croYaYJ;ys*BuEwQy^sbC0Wi_9D;4VRpS{K|P@f<`l9^Mjjie+pzruxz3t|`zi&8 z$up8l5>v}2XE^8I3rW?X?^W=hdxbae5NHlVum1(Qf2Bgs^3MjYp7Q^JuKS(F`Jw?8 z#jY9*NxOQlVAITw!Ml;WLpX-$tl-=4P`1~Sa3E{S`+fwANBU5e9i*|>Rl}6At|Fl% zv) zGA>0?>Fu^6fQuu0WWAXnU+%A38ett~BV6?Fx73`Ga!$$zUJY%H2g@p(sX-vsbh^y(T)FsPRja8>u2fJ7?Uu)uGj;6AV> zi4T~hsFsqW@A!)h)yw`DY&iebwqoV@$F@?|`+I<)`aWy?t~WuIvsHm1)+`}3A((nH zieqng6P6QhLil`+YOle86Xm2h-tv~U*Hkqt=|~7BP>;eEUvQJw#(Y?YCZynWr_PTa zlaFS(GnnTfhm2Ys>0;xHchvAdVlbHXoQK7z)~)^{Z}9uhV@*Y#w5g153l@?6u%gOg z5Yo3qTW*1skn{sk5;ngxMx6Ers(|urPqrqXGk&1Clv`dqqMHxFmPk3qY!GYASwIH+ z3&{}2HIy)BYeKTaod!m5JnSA+uTq=u1T{9L-5H*wi`{0s{Gr`{fhQ72> zb#AIvY9GnOVILhJeAtgY@uRPI{`&n*Ap6V{Gwe$}w5x^B^h;xTmLhAGSt=i+t*gfXM?WlW;p^R0Qg zhuoTkp}Vj5x5Ky`3{lFhBrd(SwbYnei)R}HbGy~RZhY*Us$!Ss*TN~`-@-X_$Ky9Oj(IgY^bP7*xjGYhj7o>UD64U2(k9?`OgveYLM zKbw>x0G|xKB{ecdJtf0u9u-f#^a~k2v8XSiSo7#qYlLDpf~kcT zV<~kKPKR%hbhu7o)c&^rs=pWIg%IAZ&dG%s1t;+HD^eeHX^?d|qTdpD0;^;Gp8#?F zD@{OVcIJNor1*aiw_}Z|`bp)Fi(FAe9U)EkwC3&y^rU%kYh+{Imun93QzHoG8}lPR zeSO81V{K@d{y3#7F_`Z7t-WINK}2rcU{WPmUB*t5-E~PesEN=s25w@|v^&4@sAmQ1 z5L;9U5*%xSQ3Ko~3bxVB=1BeklU}EqDDiN2-Mme+2F{fRjd(LqLPu66t;W8By^6tB z*C)H|ig*xQd$V(j%c@5diDY!)#AAK>4OwOjqlM&x0%yU*kQ(PPY7H)-@$bu|nB~U{ zp_oqjilk2XLDCsK>V>faPI?9WoCUZNeXHZDOdHZ#(KXv>mm`;ZfBj3Sh+50$1)9lT z0mm)x!2c3Trhju5F?0R%7oRX`KS+ou_UtoEtD=@b5TT;hKuac1;yM4Y{~H7{T7?t` zkv!_d7w*tpiBu5cXa_5_*#0R-9Y0aQLNp4WyQE}rek)tBZBNZA^$%gn4a=|^>{dgu z-|^EJM6juQRo4CLean%F!^7|9tqgtk+-GUp*;0`fpa?uFdH1OJ*{zGs_Y)^efJHtn zmf#&98D@9qpfSxfSl+;TSkg&y&%l9h>uqtj*5+ZrhR8$(^@f>hAtSQHBp7#1?brkW z8!b7^N)@*G!`&eMY$1Z+N*K_SS-1$u(THvl%!;;@Ya;d{+Ca36&^Jk4XZUvUM{V?r zG7^)Xe=nn>v``wlB+H|dM74&8^MBCU%Ffcq_a}Q|~ALGS9nh%e}z8IWL4z(9C+q?Es5=$?QE!gh9p37=bi~U?HvK zk3WLL+#U&EhyM^QxvT+EeN=dOF!*)l8UmLD30tG;b$n8=Xry5C-zjgd0aE;-ex$li#e!--Pdl zvH@xr3SD3~=SD=J)gwh@MLv?1uLA75GC-@bVppL5BK6eDg3IznZk-vPZGN4DepB?; z#F#jZoTT#$5-Hs27;Vd1kqzWSZ^&p13v3KhrO8A3;pzdm>c~TD$ggZ{sx6ungy2<)(PD$(ihSJ6=eq1S8&;3xf7n+&wV#klvz#KLR zajp<-@iJ1DXoQk_qy2r64rS8cloyz&jaXxy?pcktwD z4_Z{k{a^~kz&2NG`pvhMp0s59qRj&V%(DBIwpa!bGwDsRW5*O~F^cm;Ne$+VNcNYc zdycT@#7_5<@*L%621(8eUJ-q>Rske@O;G(D?P61a>t-cgqA8?Y z&qT_@(lWxY$~Z#a7*sVHE-4PC>YH(r&L;Z>5ii9B0$9!<^WODZSDBSmELa}3b8_h! zf@|xta*5w)5BsLPKokc?+2VgFE-t69_+Pfl12n?m( zSNAwb+($vT4cP?)CC8)#G@NMrS+{6E9>A_L$Y9!W5ohBpv*g4RsdSR9l?y|EC zfx-f|8k@RsPWrJzJ{Bwi!DqsWM7&2~F|j_AJSjlD2l-ATrM8X|TVmAq2_AZ5r4;dz zb$>5ys=R#QTXS-Avr9fnEti`X59d|oC%XW=tXv_sMFoOmg{D?S!+T+K6`F+B;eh>LVSV9{SR>7}P4|bqXIJg8w ziUh65`Qf&HZRg#M5V7K!5_1(^DoF;SA}a9GCgC4LHQNs2Oz_yw?75p`W#Q)w=_Jp? z6Fqxf2wqflxc$3cVkSFzVIuFK3fdNNRppd4Mpc7trK$fgrbIhghi(V^ZasqF!n46{ zNfBWA;N<--@-Amg?#7ubx&Tm-gWz3xDC&Nsc)I*B3LAmIZ(LO%K<}dODdc+IV6Pt5 z3-zG!2}F!ObQ+0iS%*9Qys2@gOED|^V3)t~Bp*-}t@ve99erj559b5#E|4Lze%=oe zpIOTxgl^#Ag8bn-CVIl6t?v#Ijw&#DE%`81PUQT7)VQEY-pkAKY8Fp{FUiU>j;b{u z)Mm553F7X~5e=7~MT?#El@FHk{8H{1tzJSIYW=aVk2SISjUc~eyNbrhOwdf)6-SbSyq(*jzEwr z%fsZjW)p*gyMr5%I!43W1i5z;%d!8Q7(+m~zjkPP-V%|6PdKR`l)$uzLNjQ>AvN?V>?v%UXh9v#aA4w#@bYkS55d^sM3FFj&(ER+ z>r2njBio;$kJ`z7hI{3lU({Ye$(h>sVPRnc4 zc^(876o^ZG@VmM0r>%5_T%QieD*)M{{L@c!OHQ^kp^E)C=V2~oms`qXzWS=dnP&O;_|^k zf5CfMTU1+N(3WKTwY1JjioVqXiQOWu3qkwlFS+ZcQ#lwuPG~VYXG+v|T?I`=r)9mT zdyhf*-BJ1YtH+ZxirFci`8lAK* zW(Mz(isT%y3@OiqSBycGK?EjgR-ZiDXkjB;WlnzTs$5!nB>!ye?w9L0F#*yfp~kBy zT$s{eniK>Xl4W%>;osu1&Y80CLe$fPGcyT!C^yQ^PdHRji7J3MR^R!_buKYaD<1dL z!_sRH+ON~~MhgKfE`Bus_wtvaLLXmgb-xed?bwU1?29f^U+KH4mzOkVmO~A(orySu zl*GJ5i|{Qy78a)?=~JK_{;!nekBr-1U+bA91pxsbeyKFv@6FOY@4MIa-kjIP3Y7c< zGS%Df*fRlh(**`0c9o~>i|2;dhT(>i2DApzb!;tsRG3d_uOAD7Dg|MLr~~;D`Brm9 zb8siu@sQiWUl?x!Co81uq?QZClvt=*e^wg|6cWWJO-u6$DA>g9vVFcy`8$Mqw7*h} za|$7I5n8vGJ)XOt`<_*4ysy!W^&?&neXX8DuKf4>+$OeFCVp*2Ze#I8PKjV=rYGER zki-LIF(Zu0!^UxF91@-TqD-ir@Es+7+Bh@m1*1@K!0O0-77K(!aZlcqtDq_&~$Wjc`6yl_I zG~RV6ERZdK#Q4_b!!DCyNM@Nds4r3;r?pLPP1T&~fu09K;(fVzyZj{fq#<)D^#rVJ zH_DZS64u$S-_lr{cco6==Fu{8B)h*n-qn57b(H9sA`nWP@qsBojvL11d6*jN3a7)5 zv09eJj_*1n1A@^M~3)OK*4gIbo85FBWjikr$yqkZe1oglE;- zSVGTh&%iPD^l(-3EPVjF;^1z6b2bgUDIKdR1Pqs|Brs>O zTNEgzLawL-sBE3dB_gl5UpcOxT!&S;`zQ$VefxBTUHi?xCHNC@KN6TIUv-%rjQuUq z5aLfh9X2ko8&IEiXMGM;p9Wt;=KPy@Tjmvq3 z2CO)yg;St#0(DDy8qP9s?Z?z6>m&K!vDEJ#3xZQB6FO5TXTORXQ1;@ff>cWby-~+@ zO(czePqr_bJwF3D=v;`@K@57Jwq`2%e9$1V1Vpo$e1nz9=DXdDHFM-aPa2W}2>9)35S&ui=#@oTr@ zrA;1EEm6#v_RBcMg4I5WzURG`u*%S~wMzvvn5@a{45cS052>k{<-{-F?)tT9#i!bn zgyviuFFMx{da_*boqS-(EA@ZTC_60)ZaCJgiH$TnUD`S*gxklF0&Cc~V0t{tL;T^Z z1gfQaa)xNbSIvilUpGb;#uc=GY9M&a7=(&y0DYm=;xwuk8DkXE01t^oXH3gB9k5? zaJ+@-v;lfDI^8fSqrf<_<}gd5CeZ(dP~XpE{HzeZJzlBeQE^vAo_HK-&;Pqx-1kaa zA*N-+p*G&SpJ{+BLqca!dwJQmYFtKceCcw4GW7^AlPN^dnKSArW919C%u|_PA2XmP zI}Z50@0mI-$%gK^&p5d&-5g}CF32&3g;ULlc8ZFow(fVzDQ$2L8xK^Qoc^pbSGs1g zCej+{_5&BMSpZPHK~h0Mls0;cP!JMgK~{9k(dHQ=R?Ga0uiK_SRYt*t3V66^>tnFNpbC=29IP3NuP$uyHX;4jZ=L4&<{;W?N3{G?2;G&60!$3!0 zFB8`ByTwrmAGSUuZ{%A1HM-26O?8Xt7+`m_u@MB2F(qR+bp6e_6aq~3!kr~wkyXrU zx!}d428OzNHt1FBg$;_pFy!Qnk5OgsJKAWS6Iblq%rsp>bknmx`q#OJ4LTdDR()E> zPa1PKn=q7jw{mE8r)CkSi|*-z|zE$+YZTU zcvezBndk1He>tkFE4Kyfy81q{tyvxkSNUzXZk>^PXycJrGBTB$pG5rvpL`eFN z#Q|2Pz<6iuEB1QiYmAD478g3(m?U>`Vg2u@Onx_uRnu|%?u z=;&Z<=gyuu=3z0P5r7D)<9h#?n zx#@O8vmmp>oN~*nR=dc`uTIoy7ht;Af#^3`4`7`3Ucl5j z6{dU&{q@Bbl0)(EvqX@HAsOt0hc8>$p*h1{FpUQvNiVW+Nrpm}O6GJbzW|`{tAv^t z==qFvfDgMTpq;h_s_Hnk42em~3G-~OVMjWKFR8C=1OxPQrWa$vSRvMd+z=*O>Ptbx zvTQZaX*@Kd)ks2`Pw`}YcyfZXoFqRSUwB%-F%`>xF83qpPpU+h>;8m|B8$X}X8Rc* z*`TT>Q~cB{7nz_aBe0F*#6~hce|sjIYxN7;=mYO)0U@(UcblJL!($US(I1c;JHx=# z05%bvIPG!!OA9Ag5#x$>f{-$37Nw zONkpaTc?45(j4a@&La*;`OgvlwruJW88|Lk=^@4FrAFzmXJuEnR=gUH`jS&@rvl>+ zHi|US#xY1K=@=mdjnPnHxd(eGFbEC17{Fe67vGBpoFWZ3W?q1L#?N+Np1!1mM(g5> z7n?_LgYvLDP7ey%w~V-RHp7|&hN~6AxE2{}X1_`j`k&PKuH1ZC-1Ap%iD%ozEYBR7 zm~qtwFY|pskYu(6MPqq89hZBdv&GX9Du{buX9;ss z)6a2l7B%w4;mej_<5*h&oC15|^@HKeC<;{}jMjJA*%mmM1!oM0;tX`w5oV<;nUbx* zK5-~V8Pv5G2J{{gERck06$t}Vj(e0nT&TLAj?%Qt6_9tbcpG&76qb4y1E4n5LrJQ6 zjA!tP@H7@Ad&HG7^2?EUcX%ZLz8EB7FcCzLT+g4LoraGI`V_o7xB&b`H_7G9@nPec z6t~@=8=hMrUtc6z*PN@EH|(x_ZQbS)KA7cZ0!p58u(5fSJ&iR3zi}}v6ng7;W@rmRel~3 z)u;^t0XTA?KP9z&AcCJ48imN4%&$qUM`XO2oY$|%C<1L?0uH`6Q0OXT>zwixz|TAU zLWEk@CSHBs6;Aq`dJ!vM$wIoi{LP^|9lm5blwzD{K#DRI3TgMq1>u`$FqG%O=XmZO zWo_K{*?748$m^3{ZQ(q{bNUCxU5Ch;=C)x#y6n+$!$C6$WDQbSDriaa98sUQ$`3r( zEzjLWT;2P9TkW+B+hr0oynn3};6XdJA(|yR-02NE_8*%gFqPX#s7N$41UJCOe=ssn zu|3r3ZS;=q;^&udYntiUIpoLPB*UoV?j#!s*3OTpM$kwM1}13Exf}fn8{>3_Voc0H zuiogI4$#tdW9TUDZvG|1vyj;CdsSse!_3$Izyn)*NMQu*$ zsM)MuKFr`^;of3!!p$2R1N$7CvzooZR96fnMxO#B?`{I;W2)e)p?E)PR((F7Ua<$H zX!he7i-^&Y9vU_rMh)H@@A8qYF^Cz6R+*Jqm(NS<>Xz{m9pNLoEBrv4@uoZopv}pn zl}z(1oGokBb~C@l$&Df~$k~%m1e3{WMk^)Z@nuW%yP&pD&l|o!yGW1}j zb3`*A8~7$8AE6z7h@=9cwX53$lkOf>E%l}0+%b=2ugs}U1UNbv2>Ap6>eh*Us98#y z%C>21O8sRW!VSWaXeNrKl&qEtm1d^35XIK%74tKL(%F}>(cxSY1rUVE_6MEG}cb&Vc^=t;`F#M^{4|_`!!?s z(}LCE5}%K{{Bao(5+LNzgWxQXaCghK>~4S%bkb`jICbuioDw&cNkg6M?=lbJ&0T>L zM0s4ey?{S&PF)nev8pV<$KP{2ck6&n;3=nz${=378AxVfH=}3Pv>D=z;nZ?=zilBu zz?8sH`W)6gAwiHhZaPVlCwgxzV{I>6i{xDqSn~!<;-)d#&$K!nS1I7)t5^PV&Glm7 z;$GmcK(GB>cvr*cp4WTarQ^xa_FCy!b4Jj=BEbJ&s#^c93;&NE@Bb~98KWxBp#n9a zVloD%_F#g@D}Mj@q1Xg`{6E)VV*HQx=l^f%jERMXaXQ)frSaf_@y5VF!t8>C1;Owd zJ4U^q*JDucmtXInpU+7W^`@s&?Z>(IxzBv=s0D4W=3=6vpe=?+(H#1E_b)WH_7k&X zGup{%pF@eozXzh_JnsxjL$3bn>|x3U+iL`eha-BbSjw7GKABsM@1k$xJf0dnDm_!7 zBE@d*AIdFW!+wsQ)ZesPwZ`I8^xh1dK%kAuMM8#1BUNG+z0e;Vg?E0vS9T?)=ftye zFi}+!HVeL6yT!6IuyK_bPO+en%VIJJ@bi&>#4^b=zk*3+yqX-D|1ZkUsk;&`+P1M( zv2EL|*tTukwy|T|wo|cf+sO_pPU?P-=d^b3!~F&GWxcFE=NR37Zf<&VZd)xW?d>E} zla+OhK3OP9$U<5rK*ivfpYzE#85VA8wh$#v8!-iq<4ie=xAbJyf|P;EQD#W5e6(~{ z&g0U!*>vE$qGHDvay=I%hb=-YhlRIB^_QJBs2@wt7 zT!wsvoK2YhxlkX8M#LhR{gL_Dba0$^T;N}PF}YrF6F_VHXBW*w=zD9~QBal7`Lrdr zKsNLFq%>HXNFea;y@qM;J;l@|*Od8r*9gK7p9}l4lSd4h=_6Dkh?qcKRyCLkPEWj) zWl4l1so?8S<0rTl`5#ib43Gzvw=g|0^~7?q?Y77?k45zU#s2$qp9v2G_(1oT+1A!W z*F%T=6L_J&s$Ow*%{Q13Z-X`ND|KhIGI}Li;1}oL^D&qP_1oP&8Tv%BO0eLdz3Yb1 z)-_*fF)MTl{~iTch(h7S&tzi9@Z`;R=JrPKk=?cplRBxesU3@LiGeVOnk@vAK=7EO;lurl|~=t~d3hm!WI`|6RDgk4&n+D5Eo84}KfE=+>g zO1vP2v#L&^&#G2T_PQz^skX+((=h(VAhuPXpfJ>}WtB3zNryAGrF*faP;OUqCxsnL zw~>OQEiMWvewiTfSRY7~y?3{gAcqL^GV<(qC4^{!ugX)7+^|>$`f;{e50~P}3lSlz zE?aTkn>N^pKww$4Hm?^oZ`jemlSmIgLbRmb*7^fB*7I$pTBe9UQgc!kWtOL628n0v z8@iX$y~FQQix`j|HUusbht=H3>N^)EKqFUlAp}jh&el6%{Z}F7;?zsm-FTB3- zh%P^fALnPwRr4ZKdN%O%WhuEUC{jyJ$jggzQBlzQV5FvjXumOZ+FpDff5KRKZzgyW zg1~(xR*kQ!7Wm(jf(!BFy+7x<-aCH!j8eVc=STi|H9MR0y;9aY<-Q=)sD|b{D_4l0 zpP{2X-cG;08BMt3?@o+LM;M;Wjw0Gi_?mWV=GCvE*=XE7DU;&5Ww~hL$l*k{hKZdW z+F*&)FWe=pSmyE3Yy0=)**}If-eHNdcwDCW!z+07RyA>Kl;dN!tJ=35eLGL6&zTgq zwc{`r?fWfQk?OIne{#Hj28K{OGFY8Te3x?%n;{E~%3QwS*shD)l_4i*)Xn^2?e>CTa3m6-F8qwSV`Hz3WkR`sZHj*|dnf@mYi`vospC>7E=?jiwqO z`1=CXJ+`HtH1i3KEQOz_?CSLP+~Un1aON{+sK0A)wBDnno-;{9vqg<`3YaU%-aR!d zTPLEnW!A9e$s4uG;mj+?7Q_tDuzTl;TGgI@f}7!GD4Jqv-wC;XBQUMmT>*FxUr&{vy6}BUb_|I} zIsp}~y5+w9i0Of_<%8q9*E{i+zOlyd42*1OH%%Sw)>d<{W$cCbV3IUV6bq;l0loe7)z!S?qBc0_Ts6V(-(Ss9LbRt{x zSJCLOBl05~MCm!#6wYYu!H&su4+4g+AzytY6WyPBW9^i9t%qdD$)2cRLDLZ$iOcY} z>mh7F`EN)0Yo{$DGp(pfXP4?UV#>T|=Sog#7D-&3Epdx;0T>}z4Zq-3c~Wv_#;%*J z#;erDiv|i}K!yCuuOKV#92$)OGA_~cYB3quJHO?f^caO)6pO)JQc{I^v2~X*a|QgT zEED#43G{njk-W$dav4emH*+g2Y_`8W{!DIsECW-P zQuM6&Yg{?7#l_Y!g1PqKW+U)k_MnrCbnh9@A%$QKcp4C3ieqEjIU8yP)ifj*1g zy@fNugVot#YO<@-B4ELpn#iB4$9PlD?kKHGl~vW>fZ%Wy@YdG3XGcBFo9l|S1Ml#T z!9SQ{$SCk+#io3beLQDsbmSQ_?-4|h9hH|6^sPefbhSd&zV2{M(c`P*p*Omx4@+9c zhvm3pQGq6Z0Eq(c@KFlw4jZx@+GQUyt0SQ;8~u0d)Y^8v=rvK7YL{_0x(-1BM^Ayg zllw!}V9sZ+Oz7sYBQ+ zbxg}7DmpJ2<|C-|w#5^$OP(!+mSrTYYl}PwH%Z;XQ*I==2~--HY!ZE1Y4N{2;jrFX z#!BmI`qy~WR8SxfSdD}0jkllh2P_%L?(f_~(Oox5L1NeKNRmh+_@5VTXhAsH?kett zz{VZNPvDlMA;Zo;7ow`}ukEh6ynQ38id-)l+|R2G%ZmQK$58pOtFKb?YfXX3*?j?= z9m&r2YOMLY&o8n0MNglJM6SG@N7R?!en&M)D!}37tK~?I`9f)%xhZj5)`e*3qwR=B zw$FrIiUX6OdV&L{yk9+npleKIpnUxDeb#xL9LaJrvH>KpcmtO7#0h`dIonY1RjK>v zN2Pzwmyc}XOgW6oIZPcR#e^h!wR8^zI8nh#d^f|RfDH4_5H;*8-sl!+z2Y*K)(qJA zc$k3YSxk_*{>{RPeb|&$^hb0Ec%5jvo`6mHCwT!Y*#ABtsb-Dqg-mUV6rwC$p$@I? z{x6caffr`&$J#&sC&Hci49FEph7@T+h8#+?A}iVy2?!jKNlg60?3)4dZ2(~>zDUnh zQQl3kyxe-O_?VGyl=ygt@6EB=a`DgVQHvAXZ}ZR8ht z#0XTti@J=yZ;wE~jy!Nv?nKd_;fm7#Akw_1*3U6!Ka<3(a% zL(@Ppq+y-R>pe;T>8Ixt1R(8)N@Z>Y&z>^@mSJoCt=%;(7}l)W|3)yM?d7;`rA60 z`;?OAuxi%FEvjS0Aig|3HO2aX9#)Lam(|cZiHBZ`wI{@hf-kT}1H5x$3Y)-7M91Y9 zp}IfbDQ&rvXWm;6gYsFqD|ZZ9#}O6?xxjP|?$Ci;s_G$z-IHU-#lDz(4gfk+U0f*M z$AOMu4LX@0GF3Zh30kgH*J9z&q%#sP(eT-=+mcKN30AhB(bDA8P}wBFhHifh@6!2m?c(bd^7^((+#k4*yDs{2mgcDN-wRu#GDD$wzn+x^rl3L_BDlWf-c2yC z5j4F2f&-qleFHAH6$q@|?0UMDfN1 zM{bZX|C$*S63py&=Iin|{oXRlHagU%(j*y+!KZ3gSlO!vMP4@&iQvk?iHs+q*N$7A zyAp7ApIq)x7%YpbIL>GwHOxzYe@ops-bLkQ&(ZSu3S}@tCnSggbQr%S015e>DA=DY z)iJz?-7<7Agu+wcE=h~iSaf0@c%t$ws8!Sfd@7Yg_J7WM1o>HfZoGxL13uIFc~<_C zD4BrhdphqKAB9Gz6#rH^hD2WOQgmEMB$QpG|8mkqlG9t0?u~ARsPqdyk%h3fWAvcV ztgEhe3-ax}h0tWufa5@~Mz{Mt7Mr7i`lLC~cqroc*CJ4# zrmCTg)|TbP9#am{Xt6N+PWn#q4$P8o=;eoDm*t>{u3CDK|0@J{rdVe~LsVZ|ujFpW zEB^>P8S@QJu@=Uioo=Zg@j#F^3XXnw`*u|Bk}R{8QX9*Pvu(ZaUg+~j z?Ir1Cme=JRR-+^!MY=1zxt@%ZYz>I|%_Gw@*x3u3P(!AcYNOj_zxH!)c5MV zfapy7dBXVWYI|=P?~8IS2jtaAF?Yjx30R(5Q)w~{h8=iQjjbsA_B6qbSNz5x`O~)b zz^`2~+-n(2gGm4M-&zgfh6sWBd_+yoI#-V{9uI)~Op# z-QS!E5O631Y^%c86t3>4VRquog*$(o=aH#=5I>?`LS@HDg)-<41ji)kR_s9B_!RYg z5xe>3NpqrLWk(>?Xefn)pMsH}%5Khh@P@~Gsi(xkm;Jde++OF|N>$$c;nGEg zDptQlKP%Tm6KtlN?s#d3#V@8js731xXQwABL~tyrf(uju!YQl~;o4@hgL=vj;|g@~ zW@SP?t>|XAkJZiVX^F(PVFhFWJjMEjN9nGk3}uPv@*z^e9cbV8FOVKop9>lCw8Mc) zR~&0dJjIjcckdBWa{QF*(cOjr{0}SoXnVR{O~QpKuk8~jwr0{V2=MuKLsw}IB0#jF zoFrN*Tk`3&$+%~H_6$+1bF>CB<4nMJp?GoMf5)%ibsX!eWy+kE6r99a>pn}#9v9XLNHz4>VQ5du zzJbyE@Gz7ZL(qpvF++Wo8i>f^EO;@>Nh0>UK|l_BACrIag*giDYap#t07-ixZo)lC zL8aGu_@smoW&Ah1&AUKc{I%@ED=-bl-z^K;r}JkK#F!A!Ujs z=W_GxVVA>y$=!bKN~Nkgk~4&~s=$ky7k{RHMmk<2gIgg%EcmWg?}fLHV43>pZep4+ zpeDHG--K-=ntS~B8LzTE{&)~w$G|zQWX$D+?&Qe`EqPbbq(^}4rB=H( zk9L$Q!OCKzwnj9ZD0vP7H&Zh%pqYm`cmP?W{NRl`j*5(=0P)75(}H=T{Aj{TakO(L z>22JFTyEU8o;o$3a{cEA;|VY;W9#L4@#g}3npYlFH4Ha7e0@RdR=BzO^xUWLh5V$M1WUDD8-AyBwd|q))mg-CYJ|*7)Y^IF6rp;e1qU3$byjur`+0ji zp!I)|lF#dHO0F##EfIT2yyt@uS`FidU3p}F-U;nnQpHuZqDR(vWX07clUT>2li<{v zV}1{(l*vU(M51&HRaklX1ajM1_f_kIAjfOUwqU~=$3$~@l#swH5=2YfNo~C=z`HYy z8*}#XXodep*(G+eB;{$(4u~XKt6gz*+1(4dpJ}V9f4!ucHaH>3 zz#541FuL>;%5_wb!D>lY2-n3;zdlAbU<@Gu&&y)!~`6i&{?Xg!gC#-xe0Ez-rRv@ zJX6vLjmiRczFVZ3{HAXH!-;>zZ@y{N0vTAJ$xqrxPLHwnNRe}`-+=xIC{GQu+#dk`h;ChYy-TuvYgcvT}7jsb%fS%BjRm11L$&qfUYu^a!2@RaFbu6%W*TiLAH5nZ0(@Wl*- zX1+1meWg-FjYKB;dpEYfU(oW#@OW@lFIl>{QF^LU)k;XaM2Y>nrvMD#zB|b%Jn~(I z6b>1v9_Pz9A|!ET zMNsm`+Eb@$LbRr3l7=VS_|jxv4QCpJnL`XtAc+F>7)^~MeZxrPt-PFjBu(;VXv|j6KQh+M+lW48 z+w0eD{LH0hyt-wWx1oUJfk)@l!(7!40;bLKLeaI_V(Zrbt;Bm;oijw1m>ujm3l}1vA)@e0{dkxzZVIH&dlc zZ!Nnw6jt2-JM4-PnJ8_MONp5hjYp_;*ttFVVq7{$-DTXlES^0z8*BF$w=6`k%J>vc zgntPhqN`puVrEy?ze!41)(#8iz-C9$Ge&Dn8EuTCM*g+K=;vF97LujOh#&=}J%sSe zy|1ze@!B~B+D)u}YpZ_unZ4MSs$X`iDli%XFcq(N3HuV=t-flzp@;HhKD7!8&ik*x zi#l}DL^N7j5v$p`dU+;0J`Ejo^&eV}SOBD5l6J220@U-KR&X&1wK960rPNp^X_fpW z0=gBLu6<(oB@}pUwE{HX_WTW8@#%K~AMXRDn8wlf@av1@*MEPJ7b!EC#VVn|nj%j; zZ!gDh*!7Cah51m;P^I{wrrtzqtgtl$KM-e^;#|KZ{47{kUIE)V-sQgm*~rEO?ZbIe zu4SB>#Nqq_5A1R>s{O@*WX@>Gxy_wku>D>Ee~TE*hBT%LJ-DVkRD|78_^Nj1^A3B( z1Ps-ep`>x_ppEl~9x?xPrsuPoBld=fpFM=j(2Y+J^bG4~|CX^Rb4%o|@Dose_j!#_ z<+0LUFZ>I#(^aRp2+aM7T04YHt{Y~mF^-89V^cY2!+4tzJ|nyw8e8|PY6?DqS8+a9 zt=yx@=I>m-e2yZ&skCuw@2#=Vac#!2T4#xm_O=NyP5}r}Kq}*Y9dHo{0_s$SGkJViq(`gEcO`~Z* ztMd@1FPhep{R>a(TQRC&b(;OjC;n~gD?=@cdpk8uU%RF-3T0wF`sxzfgTVi65U8KI z=}b8fwEVdzSFf;r4JlpVQADYGhWb|bB0KW|dZW`E&;XmycD1L9iNHLSga3XHj`zS1nT*Nnu{IoGZc74gC_fv}zMNg#aKgW&ba_*OXrH z+U4b$z89BEo?SSiFxUss)n!P@u2WzzK7~U<9x!`Dfd@ButfKrn#SvSoTdED8Q9_YT zDtZ|&t~ZeFnd5ZxJ1~=7q-Al?M?b&K@E!9+GNRb%-@Px|4Af~Ghn2ZHpau-n=dJG1 zr^tsLYbo}OJnm#;f-ua}si&_ojFhI1T|#E3nfocqjc=lqZc#$!x&2tbxs_0wGZUN{ z<`{Dbc9X@gpc;{V{~Iw=N_(GNoknhO zI0~g+oJ-M12WrSZn3C|0ayG-;%pHV>vKsC}Ha{I*7f$I?v$UM-+2=w>UH})!_M^g2 z5pJF$t&W$IJ41fVB=yRt_fSeZs7;R_zD+YGm)aofij6l8C#N?^$;`Yy!mX|=KVBiU zre5@8)WsGRDlua7cAO=&XV`@jK09eLOdk*#85JP!Q|9=lVmP8%K^ln9S!s`MO7V|O z!y3aV6=V`7ZH#e)Q18)7#|I)p9GE&h4loGzL<5NoVVA3c@P2u!plN(mBX^Q%s5~8F zd0*Em%L(};wh!52u4OP%(ue58+On8A+27O9(c0iBll&DM+MX=_SXCDrb%yAB^dHUi zHTB5N`TR8)sRn+)npp;=NUggOQOrxKq=6qdzFv{}$e9fLzA1gF6+EuTqhzqJZ(k1s z6(!s_tQy;~n*BFmso~t^9o4O0WmR3Loi(rzDlb>1)tlhOC$YS#f9tTC#6#U}b(KRT zOd7~}{&MhX+pggyr5-#z&TTye;VB)Yx09fAYmQ0Hj<_l*rM@2b+kGx3t#;YX>mVSIc}HlBQGQL4j0lz0ZtOXz@y z#B)CpY>en@4c_%dm=9}3vCMwZx_TOylbQa<19a&IR+;h=!?i$;&pJx1ClM$9h+^OJ z`l{;R7zUR(BIR8=s1-HGUCU!k4d@vq0h zIIpWqNNLB4vR;I^YeWfg#rOFOOlQ<+2H8+x5U#s957CVxG@NvoPn-9!#DC`<%ei89 z<({&yv?kv7(J5qar{EPG>ZuAsWDB`_1n`g<^gqiooaOW*h6YY|I^dd1l4roxNCT_t9_u)Q_GA>awwEko4x%5fD><;t)Vin!+syfM@Rp z>laN52G2==Mow?GCJ371KJ}#TcLL2p>KoQh>C0a{$z*x@%B17nwViId3@#Dxy`IR6 zI{QZsW40k{D%4Up0e$NUDuhJG!*n^exSyPA<4pP*tzM8>og}_{WLggcO{$^m`}O`ZpM99-Qp&$&$8{XDCq|Cy{|Mvk zTOGGIy1Go;Ud8_dO2)aOb>+~FYxP??$Rg?xn?AkQQu?}Vn>&a1*W>Lv*9>i8bV_$l z$AleuF3B2+^D5z7Y8vZFZGP7NxcTYTG;<{C=W(ayvpp4uDl11B=l^*J^#frisnspx zX+F%WY@24v<8CcNRI2Cj)-f42JJDrpCT?lS3qyH)6E|LufXAE@Nl$J0v<%v6jyD@I zf1`W&I#A=|<4wQ~K%L2iHEesk@EX_Q4Kop?`-u$UcO#*r`fw?LHr z?venfdz}H8od0Og)GpX1KG^HS`7R1LX+2_aTG51UkjlKP6xZTtSn#59nb$9qQZ2fw zaANqL*bYVP9f5=7NHY|Mnl#3kjB&Wp){ayN?J1kUx^+dj_LsLUU^%a=Gs9*&nS()0 zU#Z-=#$bf{ADc|nUSS6Lu70x;k8s}sXsz0Xkh$a!SLyKnWw_qdvlwE~392=c=rUZhr zd8m=9gI<)l?EHd25esh6KU7x;u~|YD{cqSfRZCvh4Y}g9GHNor5)IrXquGggZlzeF z-G1FOwq6tiYBH@T_o!r#^{}V*3L@}dQ?UtX3y@snytkdRbHZ3DF)1?3h?RNCEc5h= zn7i~WR?gO{ESJV*SCWQSxjwq?x~~4xvQuZpxxLSf@qWF6x;9%ZUnp+KaS1#S4uLoy zue+aD>l@N5CMHGtsi~3MwZ;2lYc{f9;akUzt}ve7As-hx12L95KyD4v*Oy@wt*Vt^Ur|#&I`GN57Lr zsLIN;qdhuK5dO(T2HL9^8h=g9j}1a1l#6sAxQiuOf6jsU*`p+XgA{rOu48db73}a1 z6A96y>%<Q$b7GV>PLC) znMcsgDT#Ja%u~isTsk47GNb%`ipW7N7SOvbz0xUnr zEL>!NeC%atY)mhlhD{6ABP!cGvT0@CSGWZ;a)7um-7Za{*>e^sI%{SezgyW^xk9ih z3-5dT(AGjGL`kscc-fi=X^KPCnWG4hfN!AmylzxGIR!6D+1Ogmc=I9gqMo}o-6*&m zm22CTD%h@U(^?K;%Wdl1Zk?M9#|^~C4Z36dYGRhJkW#=H7p!fSMTu5*B=*+h+u5~d zn`V*gDVgC?$+YkUv+^M7P=bl5^IhN)Z3~HKB6DYl`$)MSo^cLO)X+_BP>7VbU_*al zvenGIQ`vzdy|l~>`Rzo0@$4<#!)UAuDyuZE@@+|iM3;3(eDj}d63Y5S6C+rIa~Or} zN;V0OEAcw0R0|qee4X|%YKvP=w)~h(>@`wT<<2%YIOu)5hX{&DT?1VBT`7qvyM7te z!I}Prsr9sV)E=D<=rqpS)ih2`7@g#aw)JlNYauw?%y0WZgoL7=h6c z9-+890f=HDQJ(NfwfpkzmW;Og5fyb!b_UXnIvH$%7%$nL(|6`0SGSH>_oLI>s}I^7 z`jBQH4!ZlsH^T?Q|2T`gnl@KM?gyAvWAIzfCauk$&8ym=t$2s_Jtg%B!O9cKm>e~P zuCx@ePq zKDx)k#~xpOXtG6v-KL9U6tFmVTXxOdgKF!hi&(>g$YBl2oNT#uP*uv%qWt#sf>B=|wFs+UKHnBZo%Fk`E|CKb9uHvA9zN2I~DK2)Ze;?}cJ+FB~tf8y3+s z&~(>fJ@PTSI$c=c#>7!rc5$rct6ayu+qzC|A1WY?X&&8p1&jf5B>y4-4in^f`(u1?id7c zfvy;LjVJftQU6LWLClaEdG}9?bk;ShXkQ2ubiYdbVqGz6TTD;x;oSedA%BCT>V6k-mLL8`3r_lc`Hv2V{t@n>)lR`m2Q?(UuCMaApuyU_`HwU!J3PJYcFQKM1)w!ViYn)%sc*dG8rg(EOBPSDBZ-YRt6 za0*qvDU>dsYjR}5ymiJ1C%lMe4gnk}qY%KBgGO#?0DPNp;l1?wfmn1SPE#z8+$R zfBU#e{6?!2yDl(TAFG;FH*Ics!;eC-@jq~zTsUfy>f1X8LU#mNXlql`9lyH$e7;fT z-|D0Ac80}Q$l?QJn{pstr3NH$#41KZefdqgQXx7*gQ=-pC_dM_hV4~MNn)Aurytlq8#oZIUr)ieXrk0=%)kH}$4RL0WZ?0L!RMQa~t`o!|X zyI=!*oXiXnN1A|nL~W0^ztELqtfqHFz`G6)&N?`WMAgr)vzfkiFvJ7|3$#Kt_YO>q zOFObJW@TkgmYq29mw?7nJSr4c2o^3|!|?Uu>Uj770CP-nZXyS(!zSz#Ca<8NrP-EE zcqSZ0RjieGh{)6l3EpEu?@9-Zx;_IDo&8P5*HSk%QxxXJ0@hwC_NH0<~ zKW~*u;9ivt-Y2>TKKx@YgHPu0H;YA&EM zl?E++HkL-LQ zUUnQ-1k)ZVt3y=7m{J{!o<_G;y7SqH!6oLO+k8NAQ~m0G2p1IM7D$$=rn%MNbq-yR zRz=|a1;1fkN3v(=txKt|7kceaLpDPm>MYXkMf_wH`%|r zFU7|O*Uf4g@z{0L|AJ+wwbk0+h7x-|0(k%YPdErDcKPG{9I`#$|4twak?$|T1#BQz zUPHGGQY40{Wgqqw0X1*GJWWXeXZuJIYJC6hB}8dw2MCb<>mcVE6Gc)28P+4y((~An zc%lT?uH;zSq_1Am6S^YsJg(&W47{$?e43Z+OF@1Q^la#gN(({!?JCIfr z`WeLbF|vz)94{&&5dLq$qyJF``2Vi`WBp(GBHQnO|JDPE+yDLh2QEqnPRVRUOvnua z!hi*z@qe4}Kgx#wzoSjeJpX(2tgojVn&nnQ+@h3O{G<+blVu9dZy4CB(&RAmuoMYI z0y#FQR|AU8EB6=)wyV+f9cIQ(tik8@G>G5CvS1`hvk_)oJ<}E~IcA35k@si)rc}N! z+b!pO=X!|7KCTgeVd|b9dii@plMv>JN zbo2DH^ov3q6AXm{GOXVL%gs+UkIi)D<_a|vbF-1LYG#)5HPsU{Q;c)9)VUh8Kc|KU zXJ5##Iq8P1i1i%y`qKSxHu4yLEiO71(rcI=^;<0m`tHrP#s7sl|oOcS}=W;x&xHTERio$Aj?=t5vI(f@zZ;YvC2MH zp0G@!rBG2QN?ZKLe);LV6&`!d5>CV?r&GdH#8cKO9HsWsdMUhho3)z7n$?=+n)RB0 zKPH%!nE}k=&FakZ%=*ki%}U7io28qzRp`62fE{Akq;D-!%~A+!&0zr8y|8Vxz1)M$ z1);-GSRqOW*)Nx3bzW_qcydyY6KL&te$ScR2wC=RfBUfDyl=SnDllxn(p$J(lBl9C>8( znOtVJN8n1NB++Hw>bupa>D4$|e&*MRr6Pbkam0~GIw+AXft;6WtZEV9Ne$uj$GVIe z`K|Q8RxFbeUlZ(Ml%U}TsoDO9CzEP$UFs`S#g8(oA?4W8ef`WIF5rev6?}a!EkL*y zI(4XzSa}gV+g*nF@AKcHz7f9h*9Hg`gi^h)ZaxVo2P(-H7x5AHq*(ld1Qk#eX4BKZvEy%sP4O{2&2ZjPoe0yYq@@id7_< z)1Z;bqiw`1=p^5DS{6D`2K&)MrrC|x;PCk|B+4}H`pDWY6id_H z```!=mA}M^ouX6sA?0g#cOvqs{n>!!*Y@&Z*USUnVnqky&#IrPr(x9X+dlBY5M6TM zF34-+JzMOE@ZKmyE)S{Bl}7f0?xB%i`E|?(CiV}MdD7+U(bK<;AHxa=XTk7>L&^j) zA9$!5L`HhO&+88116koP@1YZSzMwpPxrq|H>T)<{ntc{6PU3u((TYJ=JKNX zZqfi^1R}Bfx9;BjZxe%)s*w&;i2+TY^8=EBhCjP6i9q)O-9WpipdHheE_|&=eq+ea zERefzRZ^wMv;s4H!#w5eHUa+;Y+`333>ND}z%^$Ld*`k~3ew74s-R=gDgRr(4{^_< zKz1*(RRbkH=r~#;NNS49A|d`WDdFBJrNZ=A>2wUdepI~0jo2f${z}ce z>pOg&v~@sNw~Ot=+FnSgn9EF!3^0fY?Yjq13TX1K9A2&BzqRld;^yw=1|W0qB5C@l zR_}^0rRnF_;w9|kC|E%w3pdEgLhLQBr~M&ofK!A6gd9~i680il_(7-$BYO3M)c(>y zxg(fLMf@Xnh+$0mo^6iSG+V54ZrYVyZfFgAEckdVR9q1LqY(&61mtLaHJMxiT4nQs z+JT9Y`dw(8HUEL&McbFUu58odF=-oS08HW^MeH7c)|0{AJyFWPIy)Wrm^;L@9Ck_Z zUJ8t%h(OH^9wFlAvDx59NX{_dIkj@t(cM$&I^;d%NLHRrWlhpDaxlMIcU)@YWg+dN(bW06|z2};i`?kt=!6xT1KLjCDr zNFZ`a!P+sh2li1IOyHZIMoV>AOns=7daR@8Sf6tQ0}1(MoCau=roX576#J{Ftl+Pm z*y3N*5)QRJr2=ZJ9Pepvb`#=t&+WH zhqi&{50OcnOSYUkbsBt(U&nJ2RjYQKR?4Omn8PAUnbDa%V~Oz7pf&7DKZONB?l~lX zBf%5|Xw#-s+$DzuiNVnZOOLNG-Ql=FPhz$Ys@_N_Cd6Ml-z#&J{DeT1G766B>e)x_ zLvu`|a(WBbbaEnLF`|Mg;}Md{ax=1VI>^EhbDOK7qx!S^31Sza&oOEls~$g|J85g5 zG;PY|QpE35uuOF8bl=b+>{>kN^TAa2 z(1%5E8Zu!Lxp%Y;uyU9sM|4e)8dci*VC#aP(-z0I0N7 z>tQ2Z^F(t4U8tIR+MV{h-Ij|K$ZUhthx1G-f_<01AaiPkK-u&EUIKA|}{96?s+`n)Z z$`MYH;8ef69$gYg^&$YIND*!*xLNUS(7-W+J2vLFDo_)M>%S<7jj)d%{6vd@$ZlI$ z|83O9;w!Y-cTEw%nSggvv4F>nNHP+1QH{QefmN!mb#|#zil-C#9S09W+NKJpCUt8y ze7?WFR;A`7#y(~Z-CqDo&JBV&(Lfz^BD<@m15)H@kBeZj_840)&b(+0Lr-=OQs}3@ zyqK5NrR_HZbJws740SKGYmXgoK2Rg~m&@1aqZzQPufmGk7sPJ|oHydhS7mV5h*ArT z5F-HejQWZkcxql|Qh!3BBUM%nG+8pHAA3ZB7u|*(rIc=)nNk6DOft`L@=|x@@`bS7 z7+xRS|Cc##kjlsL*ld|8*o9u`%lnE!r$mGVK0kg zmFKUQH9ob&;~YfkW&0z-8Z!;5JrI~>2x*oS7#dszk!CUGvF`a*6>M&Ghu8P7x5 z>3Qmd`)FY$s3FNiC5NcJ3|PQYj#U8-$-geLU7fN1%*0%SB1bwK*D zq;#1aN2L}zuf#Qt7+`0*g(x4ttwF2(qInae^<}gO7w* zHyU=2eR;!QkK4jVE>3->@u#`a%Ma!MP6kU=;KcB^0b>RIAiTBVo@I@!v2WiZ-xkL?7b2RO)mFf6KtZf^l#l>8(t~OkDWyMZCXdKV zO;Sw5QNpMOX}Ur3+mb7|U51%_cymL&AB>biFOhQ%?4a)ei9j|Ls~{7oZerhMNLY;s%ve?gtH}B$tHirYJ>TzIBf(WyIcGx zyOYz$!>%5~AbvJ+p{d>-(#Q*1bCsi{OK}yiX)6TbDiWBmYFJE~4#;eFt&CpO+wZ_RwJ9s0jKAXp4OhPZ-^P`3y;F5Qa#3E!grJ zqVQH{2fwchhyhW5mret&yKvblF~D>@Zy<+6(6WwsY}GE9Zgd`xb`!{rupsqK$o9;r zz35h)L^r#cm&}6m#cKlV{Ez|hh!wQPz$Hrfa+!vy;~b|pdnh{jjLg!wY`Xv@RzC#0 zTNC_d2*|Ht+BQ20t3~4TI&e2 zSv0~yj)9m}Ppv{Cj!?&aE2o5n3=8Ora5I|cUzMt(Q~+3fpvzV53nF#kvhJOTGNvJl z_Pb>`8%zk?S&)c*rxPIp2Dz35IaW0_$_7F#%Y$d?x*`tW9YQQN{NL-NYGUEnN;^KI zTCaQ{V4nQoq@8`rEx302Up=h9p=nmtX}!$V)jGH>Q|JjsnM=xWj9BTp|6NSqH932o9AeO=n{dB}6 zcB30YN+v+Ny+znE`|OrS**vcXDj)EY&q&+C%|sJ)%$}8L`_c;plz`M5{5@FVSH4D& zC9;zrKpY8AT|o7sX#uZ~-ZVF(Fh9h9QL|U0epCu*67(Hfznc!7%0q5?3oYH|l z8NawvU5UboETb9FF*#wV%MV|;ix6XDL?qvltQ$aDz>L&y=!!v07^x0fgB!!$S2?bW z!N9w|Ip*zMK*eTCL<>-6W`d!Z%Q-KF{Gb8fLJNG2lZ8z!6Fr43^&j^-TPLBS+ed6V-BN?P*Fshh_VDj(1sJb5=afL1_5qi8 zhX&g4YDfX~g}f=a2tAd?etE`d=W~3#?>{M2Zz2EYIOVvN&-SKf(q$3qJj zt#=JJ^LSx5*RWS2d723nObo~23!8;a*PEpE0h6}P_Cn1haVu`j%n9(lJ%@`a!Ifr* z&WWrAlOr41^)I?gpPk)lLP27t>q6% zR|6Xl+q{!m?VZyd@P~|0saDd-f~@C)aB!F8Ga&^W13zRMy-D}5#IfKedOd&)7?S8z%^=G6B%O) z!ogSo8c#36du1lRRjRbn;NwE{Z>yBlsjxn6%D~-`7A^vtp(tshBSU> zZ0@TC-^ObCz0D`d9`BINB++Nx7SY+ymlVsDn}^p=Wof+}hs%+b1hQb8nRvh7k9r+; zqqSai=2A5YGkAnbvlj5Ts-83>ZxA`n9h_%_h>N*vuNs#yQzQSQheMyv2 zdBN0DpalpA!ArO`bU^If$v9WN8V+&Qs%gih(%dAy$|^PEagST9|z&Euiu1cb!>IL{ZaTkJYy7$$j= zj=)!zdoSRXAX5Uu)%@3Yx! z>)vLZq?HbpbO|Tf- zik;B)OSr~#dd$YCgvvu0F+l0eJ=bLC*blEnr;+fnTvK;_b(m?m;=KR%b zP4P!7w&J+eb51MH5UuwdV_Pn}Yu**ltPBU+U*Jf>I6>eolztG}8lPXALHq?*0m$Ic zkTA9sMJwIS%b!Z6RAJe0f=+0N)wL9-P(76R((XVgIY~4Xnw?B&wd7i*D4hSeRA0E@3*q+w1f90M910BnA`4 z!hN9Qt`uxu5fa*j^(BeJoR;XwCw?g5&>~hav=dg^X2VE>!%jqPzHP!P zG0rF`EWc`P%_O{EyY^%aB=IZsb9)oH-4vqO-kk=f&36k**d{L4YKldp&5?r8djVDdSU6g+s zMGpkeEcSVhIT!5`P{C3MP=ucLq-SPJlv!F2FPpm*_L!747ne?=SHtY(t0%=8JUu0m z@CQfNP_j%wix!LMD2N^ct}RuU?SFMMYrD1K4D3+71KobW_fxnGaM;7Uqv%KQVZ=Z>M%Wi_ggo8wRy&mDTj)mYgZStfT{^-t6hm|^tm-YMp; zNWmp4^GI4+lf*>7+MvCChhr@^&=JGE4-EGpc26MXaf5+MLdmnBk9D?ZfdDObSX>!o zGW?nT)j>#pGG&6tEo;KEW5jZJ+BC8Kw^q*`Y1>6yt(h9zyz=T^w87Oji-}9{YTwFA z-nie~?k_3uZ{8$)TVbecnW$2r)Q0uh&X-NnsuK=~enPJh(7 z;tsw(25NHyi)4UEQ$Ka?tf0sdHIo5UC3a_(JYU)O{nw#A-R8adY^@->%ysd8ktr~U z6~PxGSZ>1uyjhmf>{txLG~;u%f<7R^a&C2gzpfft6Lq()o*!(f(Hk_jgVzrmMe0*Z zMp9^HjBG7pRFy>fGQp@ZAaaI-&#+Et)0q<5YXkF)0kOnj8!RD}b6qI~Af8(UO5TL) z1saK&F{mLEA1dsh%FBBw7X($d&Mgf*#uZOUwlK950;gv?Gx8VQ2uQYNjm|E4L^&Yw zqFoG#a4bdutCG;4jG`^oa|>z`W&|56h>i88(!of-eks3Ec+^^v#ndC3Q2RKb!!Zj$-q|Pq`yM{1g@z{cGiMwV}P@#DBRUW@Gvl@@F>A7 z*T^8It=&AdV0v}YHh-5=LZwb+@`VI|SOH`RbwX&Lp|ff953Pilj$2M5Dwh>D zOIyG~yry(yn)rxlnF(vmfy2))x3?)EJZ2+r=zRo|Zp`|%plsp$%_aU%fhRV6l4Blv zr8MXTWId`Lu=P8+q^sm?Ek+ng=tj=TQ1wsH&8^K4%AOwYQz4U}vgd00T4^{tkRnk5 zvt9gmw{@CRuQ^M)qI}gU(|idef~gJa5PNZ@vVVFDfdelWTo;OS#1aXqY!PRM8Wa2> zBaZOsKxclM#+Yb)@b*Q$@Mlk*tWSJ-2D zU9Xe#gv)`bd_ewTwoiB4ft9Sg28KUDojxnmA^o2Lri&st%nBQ2x&ubJLm)?ath%~% zi>$h2pKXpXZxWi!mNF}b?(s!y;sR6|B?{H5iP|YlRh-xJ+1wofLVn`b?rh#(Y6D+8 zZRaPwif(9mLz!DG^rzu3ZRJw!kQpxTr>F&e1`PeQ{S;wa3%7Xe>#Jkh+?(g{bb8kI z02hm()HIqMa6t!%sRR%2rgbrAV?W02s15TllvaX2;5fmmrVav{H=C2$?d!bPi1jty zO_R2ir-GW=4sAsm;=r*dmv)vNo(}NDiPh0Le^3S89molHB8VKcz@BCjw7HU?4>Bqk zmhm7rquUN*Y$xBRsJD!n2kGndNnzy4+|&4yPy;VV0UFUoB0|cVyHz{-@tGD3Z=ZK3U$CAD01ZNMpu z#wM~+q-ko*sz{ks|53jqlqr+wp{@>SeAW+L=pTc9vWu8bVxbfADSonzHjbc;^AZkJJ50CbE*_&4=2x@T(JY6shCM3b-mHGtn zNg&PPOQ?Zyxk)W)1Dc35o7Os15=a3F^D%ftAQ z6G?C%8-l)FH4K+A2J?{B(Fih_HQZR)q??jx3CHr~8hg(&;Tl`M(czW$A39qfxz`w@ zDhz@~77C&2P1mw1_D>yf&v|@fG4j>vL9aFU*1RxOWi1$$P|@up`ix`M;bU*lBP3n~ z|H>5Q(}xlGd1}7)9d}~{hW2(B<%@qgo+4yE?z%Tyn7bov*3CECQ8xLSG~-~9C{JoaaCHcq;Qc2A zZl6fhi6!aLwsPIC2#zhy7<8hoEHc|sD+Q+vH2`N2nW43*o%bfOAEk<69J9BoUBKnc zop$t~AUSd8u5JtpzsbMfLEN(lv`e45(;<$0z&Knk>K z-DXA6nGr%^86B$-HKP_}X-X%>`g#7$Tx@#7uG!Sm8?760aHJ}U#mO{p=84pOwjNCu^k-D@#pe?`)nE%Tl+Gg40O3Uv%aXdq?j2a*VPtY z`y zy0=J~ljr{IxZ1<{Yd_I!bk9z@_KIKmqz{+SBesW+7EgNZ1x36)8`5^jf#k4wo3nOH zPw9_lm`Tcpq$}cj@p~1`--5z7yoy_xbGSiS=GqQ!Gb@#gYfSG`jI}@N0EtQmaS8AS z6Xs8Jm`xP9VhzEt*N<_W>;5Mb6A7VB`)r9 zhAb+YlS7@VNQY7qt0%Mm;YO&{ex3+?EM+QqcKG&C2YRRg37B<#{oJImCxcN?uz?@^ zf8P5$EqPU}oW!?*PvC5AtLDwAvUREVyF^H$BFVEtlAr>5#DM@WVh0Gs3Nf(qjdTrVvwcm`IC?rZkGP z6o*LW6Bm$YG&2}B7i+xR;nwK8MN_~g!U!jxi5?adkmZ6Ztr#AUnLaojjNX+Tqd_IC zK*F-wgvwMiuYzS+k4dv#Ktu`7g9o2aW(VNOfr==%O!b!<1mOpJf`p1)%TS zH!48G&&!~;D9)&IxtSJ9hdz2V*xz7!TiKO5L2@~G%GnX^3vgc)v926K>igENtta#o z$5Qed;ftWyp)3e=95|_D$!~}>Z-)fId! zf@Q0=f390rCa%U@@#8lPhjT~f^w*F^2E)U?(!I1ClsKMNf|^T;&QLKC8;oxw#K2@L z{GIdc@xg-mpI_wFx61-JxL7M83Gz8JH^lgTzk`pY4ayX>Tu;=tgWeSY9f_sjU1BZenE6VyMw@}41+{} z6%6wYq32qTA-C-UbCo0_zIid)zmpJukRxMAR({8B*3FU=be8G?jz6*kffCVYxTJm| z8Ch+i;E9SHY7Xx|ibju!BFgklbn!kKxGN0UQ)ds18;XKm!Ww0mQx&q=mM6ibjVL5q zel;cuX6~r0Y#urp*5Zv9rIw}+_lRdmGDqjjsXLO_(xt@89bMZKCBL&~3Iru8T9*@M zG1KE_mQ(s(5TXp)QZFQ#j^o@Mx~c5DrkMf#=Pe%E-~MJepNjQ!2p)nLy6WUvRnXAjFaLPjRPK3qv^mqmh{6ub z)Xr8R3pKn-5!zqDOHwyPvhb@)k%^Q$WUAU=n&huswe!8s6r8GW0+l|Ymg$7MkbQe+w7gkHt27xG%_c#B?xT?jwfFlxA7a8L125`7` z=SLZndn7Z@cMJ=24Ey0$Zn$grEwZn!Ospe7UbBcY>8qXDOf1jTT`3d&P zQznhvHpUXXEn>73zt6KE!oZwt z8q)o%1S}bx6cSz>r)#s{i2LS*gV2Ln8rB#-Xg3=?BNyULA3_|(q+O5)^9#?V-SEdx z{sj8TZ|MZk^8?~?4Q%qGeA;X~Qb1PmxZI&AS|h)hK9KY?3G_49**SQxej{RrwnD`o z|APjBk`L455&pN48MQ_br4xeo>O-yZQF#e)AYamzu^!TO)s&u&k*e=UK3z%3C$AN!tw{+a&> z7NRCTKWkiF9RchDY}p=njIER8(;cm0Sl%^3O{{SV-=- z6q&a$zIYjKtluufwXb>xGmxKW0l9_Y=W}(D1n@jgAgBOfj{u0`m0>?7IO)eKcRgKX zvjNrA6ot|e<-A>F_GG}KXTfoDp}`S_L!;ON!=hMYz1=;2p=yd*#6r|m zRY252S5x7W?#H41@TXo+)7Ah}m&(S)!1mL!7{BG3$M7AgnrQ-faSHgHd*|98J0s1v9Y*aK)SlrGv78ucG)WNJieRBD7S3Ux|# zQgvE&3Uv~78g(LdDs{3od~q~nNM4yCihe#iC|v-VnZ-aua0#fp*bB}s$xiO8^}(6L z;^-G`7op=0hf>#TO}X24m)RDNvf_O=-Ok7CUBsbZkKMOV@0K6Outb#)zRLZ5kOe#( z@8g6}|B^^~Sn;Gu?PLwm<{$;U6>k0b=@^D1#+${)g$3HrQ4YPuz*r#&(gRs__M_Gl z!D}OQ3!Y>dFLI_kdA+B06sCr2B%%OZc2ZvY98vJ9FgYwveBbLZ%}aKxaHbgFU^=8; zq@kU}L9YBKC5Ni=H<0&!>P4p;Zdjrhm2B~+V}k+P((}zXYRsmK5!#&qb4!Ibuvvep zZD0BgCAiB|5y4bPCi~sf-sVTmczCZ4U&CGX8yBMr24`+T>F6p#(Xnm^CpxIjzh79Y z)e{1sa^2GzV59{znzV1(lbU^bw>G71`LtTPh0IIUOo{o9+Mef$-5y@1hvd>+!I!*Y zb11}5oEWl*A3%1#0~t9Hpquvy84K{LN*>zl-+vFn{xa6%JJ=pk>yTwf`r6F<7) zb`Ba}|M`e0MbrOs5Q}&oC2s2?oOi;OhJ(8^v}oVQ60++ytI1;LYa>@So?RhNXABDm zarK(wWKatUIP*0Wswre_=^I}DYpnMz;-;dT4xHv8pA+ahpNv_nKx;dTUbJElSAvU~ zN=-Z^t1wGgc_&g}2*JDI4TctS>&w6hxUpsjs0u7zzIXy#?FN zlDs1jln1>$4R(G=?(8TRQP5HSD_Ywamn|4kC)#K>H!cyfExK+PAY-YMU4V{v8uZTx zH{NOk;Q_L^FNnNuoYyrnv3cfm5o>9kOTrUpRM)qCJ{57nW zf2>IO=^{)i5GyMXThQ6cmJdqqT{*nEA=2;&AVZTCQaTmCVm4_VtD=n@Jv&iv#h&5& z{jkrfPVgc2&Xsu7qCNf*5Q1@0abjFU;Fe;maN4g3^J>w~!N535%f z!iFwMcErFy8{W2uLR&9Vix^C&lA*{hc@Jqvc=l}o9J^%N}|FjC4K&aR5+{Y%<8N`R=eB~x{y)i`8>grzVZzk$Gp6qT;1 zqD{-B3I0EE{h~hu=gG*8d&x2_(n9`tQIrx4tKHJb2pI%OfCgWDzcQhtoxpB52u6~wNnD>D} zms1U(JxbtrUfkMbT}9mt$2H!}78NvM3H>dRrj6rgda{bwYV6a-KYDSqQ{!r|;3+6#*d zDy;6WLFB1BpZT01_xZE447v07jjwQy585N17CYdGN7G{A(J4{nyT36QL{vy8!pieM zB5~7j)4Xru^9%|yu2_zY2me*hF~EUaRY=P@w^yv%{`XE^!DX0U9;5iH$|dgXK2hmf~eF}d$!cs2P8Q7X&uy|;Eeg^BSk@{~ zH0Ttzb5xcO=TcK^({xQ=b;OuRFQi+q6e+tI?2dR_q!w<=o*4q-M+3=$P&Kwmd(?Rq zQ5n7TVwueS+!GpUe`uP5?)}Tk$c%Sv3036QU0Acn|5XlKQR%>EWj!EdLqu)~UP-kf zZ~S1Qz4Eh)sSU*T?t#& zKQb7qB!piFJ@CF)kj)9aXJgMo43F23v#WT07>F?ke5pwugrn;wQwwVa*QTnKweDl9 zTX40ctUynv^=$X`Mta@Y@uV5>U`Hk%SB-MxywGYTnO$t45brnEGwxsO%jCg2hi9dQ zXe=y~A7LZgfrtLKT+VPQDS}c3j4WGWf%$JkWr$20=mbP$MOOs>b(=>~C&m8#9^^$2 ze}FzhFi-A$Kw0H@CKwdx)=smT;mS@w`*(M6eYl6|zU3v@uRK zyB7&KS#0jzvqNgl^*x!4{oUQj;3|xVbu!9Pc*9fAA;LYhGGyX%n&UUF4E-noX+KCF zU!$6(G9{q;u6fFt+sw%OkZ;E4e%ZGxW!8af)X0Wctf{5q`&Z1sSN+uh)4t{|hCJDQ z)H5NOFSr#bQQ3xM_u~ep&jRS8Ix_Ozh`3iDuY#V;F*r_oRv7DZ%w=>*9z<_C=0|87 zs9yw5igtw^wCkaNn;1x7G#*H@1s03h4C#Zpcy+7L#3Jn+5oaIr^$<|H??qJHtD&gI zN|H+6)PNJRgwd2!SDIVBZ!C3{4F$DoDhlNVo6qa>ehZS&eSpw>?!_(&+2A5N8pxHs z9CJS(V;m&c9*RVFB%}LyiQwy>jm!DS%C2OU<2I<#WyH{MaG~JAekkli*=!rEtJ|tD zk~jOVoNQF>*JF=qSj&bg5FR)u8Wt9UvA%=t?E`ToD;u)Ei0wFQAU@IA1#be4xK=ah_2Ac%atPbdx+^)VaJy#C+@%<^dG!}j;~a4&wwB6|GY zUHDjsa7Zc9q%?)UtzZ06=l#{1ivhw)w#m?~XOdKs^gGi=M@2D4izmR>ve!Y8LhOsW z*DVF%5If+KA|4qd`Iu8caFm)w*aU?XF>8DdsU2vAC$PyIbzppqq-gY(_WJgUpSQel zF&~H}%%n1q;1m;bh{H*{kr9h$&Tn<6)!pOK3AoLHh5OVGbF%g{>g{?XfPfifLtB?T2O^ zpLZLvqlQWA=)VWB71}A?8*+#>_oC7^G!Cmxsfx&KvrS<#E3B}J>nZBtw|n%Pb%;b7 zv?%vgyBo}T$ou1QO;RNDLmcFF_cn)p9NWom#boXr$m;rjy%iwZZ0llEK_m^iLc*_r zYGnwRb^ZBvheivejTMaWS1%sCG$WX|YqjQVF3+$`E*VuGpqJ@n_P5WI*kmW_+a6_< zglSW(!@WXtVNR!%PMWdpyZfU!pM7ib80Tp0&5hst^9>)g=Oz#)aBJmejX?~By90|C zKSim8YSRaBq3%t>^!k%((}R1x;N|di<}$QM58wl|>TkELFF0`u4nP>vJ+S0hrpiik z!^pI5fRG@J>=K1o)ota>%8tclFTf$K` zITJUe@%)QgwjryMIjER~rUd!LTA~UAiM!pHy$Hm_bYAV61gfLY@0>$5AdF$B(`op_ zlVpE5xRw%&xKC3JN|_IOJ+y7giwSWfl6Y73iKrREm}B{NQC@u16{+Xo!e1}N_%V!R zKfov3N3rWIbiLi#didKvVzi_*zxuC%GQe~%*@?kHe*1RtZ%jXTRpG3!Ox9>)iFm`Z ze6taYJb-$s_hH81a`n*(jcU$Xj-2v;>Fs**L5vRzt?;%C((>B;-R;P%wQXK+1LI~6r;oJ zy+LeVX%mp;3)}LbV|^;mX6tR_s{4t$=`nhmK&WKm!J4B*{n*=o(`O3q>QKsy_xzKu z*jBQ6yG1%Nx119mmue7izt`qu0)avP0sfDrLP}j`pT^y*Jq(2l{^m*8A|-(_4XduW zy;fA*e&fv2f55%-k)CvX7$Tt@hnSXgV;)nf6~z}TQaqj<_0_Zuq9xa@qRcogYd=*cHF|8XEdw%|uN zHn>GUU+mu(mlm4Jb4TZ*+{Hs%w|3A7)edhf(+AGm=jylg*HiRi=5u`Jv)$Pzzbmun z<&_^eLj5+SSykIOJ&SY&Qtt(R#O%jDG(@ZgOe_doWRNMkWJS|v4Yr&mb)r;D7y0ETShqZZt*BbJ z)pA8&tn`@8z3KRJL_z>DfD@#S!F3DK`~vYYTvuST@`@X$siecaCu1=1TCvP!cf4P> zzV)!|hir>^zDcX;cA3%TPEotb2vklbY5{I$7U6m#Hi#w7^&1@{o1O8lK*z#@TTN~t z)v$R0Tw&HtRESGM3EgA5J&~m@X%)e_r~|udkc6mUDGy%O*lGwDV#UzEW$iNAeWI=$ z#srL~-kh z4>ocnsEHxl!`D+OO=!SE04nu`$IP;R(80@+#psM#s&~{RoNVBeApPVSf{{9+MPu}K5$yvptR;A){lM)_i0RPx2$qQ~E{NVJ}ho);k znZ64?we9cKb}Rw)A*aHrABav%sj_9PN$0z3+*AT12-vmyxUy^-3dV$y12%<$6Sc~T zI7k|`GHO+qn7LGWs@2pePX%YB#v=9f z##Q0QzmlR%^SmF|PHF%pldE<)n?jrx1}nKyZ<^^dR`6M`=aHVh*t1&hn)Gjq7SuN6 zlGY7XT9pv0Sf@|d0Wa+XMNH1<6Bn+L9w`nRK%PGY6Iz z*YZTb-EFwUdW_Elc<=-FJHb6vb=Q)F%u(XR2EEJ1`Tw*)?dUuYoEfZG(}?2m>aKKEoo-mpr}=&-+suNVJMMbv!{D9n19*RqeX%+qN>lekyQx^$%+63KP98|8`kN+y?=7w z7xz{PzjP69Zj}G+e*JR8!%6ZoH+Tcwzfq=d*umhu-5b?v&D4`bGwaF0N!rfF8v z_$ZN-IWD<+jNe+Yv!y6)l0Dxx)k5dp5o$x-Eut338Dg_z4hbxIT#!j5uk7A#+tvO3 zxDEwGr*rL)PY|lvEnXdkEGVDS&njNSUm#WL}WwikXE+8i9>%(uc#WGv&PpH({D-yb=h)MU`>wAAt3 zbYb(NOT{z4FIBAn&5y#4xqRt@z2SyP$Q2F)H|iBJsN(~&!5fzgo*mR9iOshXY%+d2 zKRfHQcK9%7l(yJ}5s)aBj9=B^c7I@$4xY^>fbWNR?q9FY4fpG_zU+J2kAm#B?qbeG z?04Y#x{a{PTU0Y}jyuj$+lD>AuCm$<2d z522RM-=v0kWt?+6#ymA3$V$JO_H$iq?ze$7XnLh?les+*w4$|fH%&_*RyR`9#Fux0IVgNs1T>rVI*0?B_J| z253>fAlt=PYb&tweV+uDgQSif**!eJCf~jgIW&zNp+x{*@SM4D()3@p%J}Bk=e8)W zY(Cy|#rU~0^)hOFo4vs#e%Ds6wU=R|Tl>-pt;}`AL6bL=0PaOiM(+yAc+rVJW?p>X zBCK-_^%tdJN@um`TJLV(*4XnaQyv}|YqOaq5&jqyM}~1>tvod$T{P1w=tOJSnf`tU zPAKC)*E$VmM&c`y^xdhAnpWe7z>!Q7e-z2db}UrB2F@HRdh``p@^kOF0}W! zJCHa3nWw*PyCm(_Frg-e`JtTRKcr-%Fve_MA069=sYgoK4(Qn+b&-)RgFM6eggZqxSikbaYc!};eAC=Ub9@D9IKJ8pU(%FM>nPmx)nOY zYIM&-!b(d>O?b80D6FDReO`S>hP3c*q!GitVctpC(Od-aSTQ7{#AhO z?;jU-$;Kx2$X({1PJw?DP@+;*JRLW7u$nGlyv|>xI{3J3+J1ABDxK~PKbaff@6Q|H z6b$RPcM*8DkAeQ91yq!2YLA>8cwZvs#qL!s;3BwHvlAbK2t ziDl^GHZ}4~EpFg-8uBQr(zv`rHK|H3-O(G3ET9VRLvR~fN=KG1Bg1hm$*lqQz}ae(e2komoU-w0P|0n*yqp43 zTvGtpTCZiNknzG`SJwYx232%`#!*Mo zsESlY22R{Qe)-S>zB-bt({>jcdZ0eV?}inG7jt5zmv9D}2xhU}FadML648)o>>e`D zn?e@ZjIxSncc~cz@GTU1T;_7n-qw)6mE56wp9U)|mrHVdefLlar~QrS#`TgktRLne zH*3h}SJGA>o+kf9@-&B~*VO(VS#h!{6({Wa$UIS^X^1P=ZFV1FP86|iRWmYu7)keO z0uO%x+RfjGyzQ_dZ8kI-M`U!8{6-{|(>{SzSqbX>0mtKQ40Bcx)Vbz#?$s@E!?dK; z`G`L~7=Y>IBke+`_Io^0xoGYiU|7~or)sM`B;uHNLNSHbvDQ8&8yu}a;`MTTTTdG7 zq}2{y!Z^U!hWATc!nqU06lWvR=Jf-|o0nn2`}1>siX3jr-qVspTq34o>f9oHl3psx>j~V6Odt zmID7>6#VU!t23p!!5EB%HwecUU-vshr)=)j8$Dle@3t3rKd$y0>%&+zW8p6~lKck7 z&_h{i&^

hyE>bxXB2}j24aIaq z?UswBkj5*U``VsYlo`Btg)Ilyy@!VeR2?aY9Esxj_Dc7}ET%@wAnVHdHC2jQNj<5&P=T|H)FUQ{m9&2+o*eC8@loNvKf`VElTItQO@Qx+?N;WdpN;?M*A6ooYei6-~ z#Bi$5H%f2;v#!}#=VcXiE6V4{s4GXyakM_oS~z4p8YVt0Pn%0oT<`7Ks#bl=f$`(cmuzx->AJ@@|RTqKRb%bv<8DxSTOem`(2A_An5sXl%`m^gzp#l}HKe)^ zL!)WPvK)Xw7Bu_$=d$bbJ;ngMU@*hr!AUtfO|B0v4Dr8cGAkW=UZ%P;Q#4V{Z) zp<;`??#FaTg)=4_e)AKwv6s)Hc>-+&0+TNmUZI4V`#gw-?U3j`4*wMSNgo37W=9Z1 zR|5|yeA&7>YRGcpq|_*y&K{A0`AD`}uVOBp;zrt9asEOPx z9RCIK0RkBS3~rW=*kWRQ|2;><|6gDpC+;_zbx*-=p^w-1)D-#@$;dcJA6{Ve{ag8e z}thC%{kUKqI@0t-7Ai7 zFJL(;&6=6x9x2kL*{QABAu10LspT%`4r9NmBk#qu=_(K3sOlO{j%DDA#ef$}#9jA5 z($~ayAD&0dS)qk)fLHk^g9FXWYgc|Katn`P7WCi>Xu|_3qqczZfY(uiWpfGa={tRm zWX4XFlOb|W%oo(DC5nXIc_R4ic9-~ev$;)_{ikncOP}1~4te?SDd*#8_gj^kEA+dSgEFcVbz#1hoTf5;*CqpR@`@*CgM zW&^xX0DW-j5PW1R^{`g^;qbbip|XJ#Iw=Kx!m(3n5(=#l;BX88Ycrb>rx*(@opsOw zF#MM2wBdC!F{67&C>CRk;- z^cS66E-aR)#Ms>6oqoWWcK!9DL;RnTK}G<|Q0!uK_l1c9Nuc(%3hZ-}o67#Q#C_+n z)c5i(jJdQpzJwJ+edTNgVZw@g7fcxSfm`#ih|Bfmz01WXovQc8`_kuInkRiP&S>`KR{F!I02OQ(atD1(%CB;Nia6sv76nvC1oJ+JD~^Lm z;9^6uOhelP(u{e0tu!u6a^VLOGe7GMX8l*@LsGq5Au9WfLe^>{BdRDYn*2G2pcuM( z-m+;WRAzo9h+HM>2=DGRwzej3$|qb)osmUVJdWai<5{ZeSVw_o+yepO6P2SMY?n(d zQ8!t0-Ibod|F)~)|AHXui3K%iO$!cVD!G^Dk9VwZq3Q+YmNr`c%)2B8da0zDmC~tO z_>u^L7ys5S{|LGK@4FHBUt=`QDHMD90LCGxsF+BVLrCNQR`(yY;QzJ_`GJD}rTYW* zV4AP@NV}0;6R5(ssc_nyTvJzQT~?}_B$^lY^XEv1;^Ge18!(7i2#6G=rLk(0=f@9r zIVu!+Z< zp)m_7N=kz8-i(HTnXIr)9?5aDS31hu_2byLqE*}4TA|A9pHfR+o{b0(LwrDaV-boH z;qYlis0f6sC@n|6sdM20gZS?oVlm7xiqNs2fj0_fb-Pn22mJTrB}f|5BjAa%Nv~#vxbq@!EMTQep7{VIIpjNp!4yd$r+X?8<&uf z438q>9NVPJiQ-pD5h^8N)fM=uVbzu;`URz*Y7fl@h!j&%zRK$I>S0BoMEZ3{_!JVv zqj5%RGS?7sa-yI8{O;tw_7uyOJdD=J2mPx0QCg1Eu!023rZ8{H6v0=NI^tx*87cea z;bbizf}@>{Ui%^F2)9s!w~@*a=NrCD2nZTEX~c* zDRvuKz$>aF3hGS-uF8elF{m^*!+)8Gih`xuq zm$R0KCg8`m9XpB@Ty7oPU`gLZ>L&Xk=|b;w>=kKCxp~95#h-l4%x@;uBH>LV(;<%^ z>OC~TxpP_sZqK25o{2w?Ty~t6V=_AOP1E%dxk+@FNkQm#9 zbN%3^bx>*HY+83A2y(RB(v)nHFKj(y)8{Vz%b`5;gzJ^D(#42v(zp7<$$xw|bVa#k7g(q4?H=t14n&AzHRV!0`OhJl)b@)roB? z)3ga4HMY-nty8cu(w|lthJ4&cU8pXC0goVOtlA{KLhWj4Ub-3ye17^+bcSm%=SQDzuh zt$)Z~V>qzMJ!?i(sq~w;17i#+o4Xz7ZL*0X8x{)L{C?v?HMnZF@f;Pma_F0>B;g^& z0Bf(PJHpw#E}h%~TLqn~ynxB_WVW5cq}n@aRUVDRfn~8HrIzYN&`S$PcL@Dq$*2MARfdh z_-%(s4Ti=^XRx@_>0<8gYblI+w6W|XZ&5LI=NqUyvsYb_D5k-~ynwK)cOkT}Nv>?Q zSU;Ln+uxn*xW2LAI#VLO{|LT`j}ognww!y8I7@_C2(AMg?Txml%`AtT10|=(Ap80v z3Q__IVU5%w+bjUMir-O0r2U1R5p652aKlU64S5#FM!$C;?gb$_NPp7G;gfU5yooTU z@hLSYTW!YLekAH2#C&+Cz3vyw4Q0pUcI~cJocjP)WL2)u(y%vQ3<{msjXW^b*N^U; z(+6zWei!q+ym_zYl^;|#6$7Di#Qxd6BIPmyTIX%R-wr{)`wn~UPoXW@+4R>?FCkTX zrbmz&Uo~&#*sDETpUB!z1Qr!Q3>m)S>VCSILl#^Y*lQfUY+uQ^Am5kf(b$^D3-aEu z1@|a~MSMSQejc`WK|qGNvO0c;?I7R?=s-m_bGr}$;qixCed`S0N37nSXf1YEDgW8; zUJ#%}L|$T6dJ6A2P~Ef+yzg)iR;fJ33#&oYYbx+lN(Z{8bhR~0H4z#qxP+_er7y;N z%0C;R@_qm|eV?zb_RQOIVg6!bX>eYXgWJC}`0i=&xOiXCA>BL1u<_CvL}dLkH&r{!FvFrprU$ZDb(i@pEU~ct$sFAX9t8n zw;5XWraj5`Py6QpS(WCU?_Tc^*shIR5wmV%K2tO%bb4;?*5HJ@Z`F4qdk=R^&TLID zK}IHISsQwf?0z4{+HXy^e1{X+0(e6vO#X^Jo32@bh@imYcg_*MVu!J4f$j2a-P6x; zBT&q~54}YlGB!GfQ&HZV@9#osDLjkZG-~JcMwVQv-h7`-OS{Uaq_&^)p3NT>sV3pl zVE+r>byiM9zq$6CBk+sR2^U4M&#UM!C6+HnY}tO9y$CHFG_-11CksSJx9J`d+|X+v z6_}mecFz+3ZI%&V&byG`>m=0qjN@%C-}{P+-r#M{%D!(#`@LFx&opGuNitWE7zGN> z=h?xt@|6;Lz@0f@`n`C5)k|0Q2~gxF0IA@yVU4X{D<#LXmvOvpO)rTRGomTDW6B3h zM?R{qpGionsTMS`Fzy?ZPkL(y)_g#UzpHcM3^t5vwkCx+dQuH%=6W?OGO)pOH_YEE zz9Y*Vc*IwIpi}<0-*eNcf*}!O7g*oB+vEZ5x7wsgLacFID;fkMhy+cemlw zuylOw<-(`V02Y3hLY{BE7(fuf{~(G4_yvUi`NeHRBj4pg!l0Adg)(_D2+CE^h{vye z%K_zgRo3%L6k<}$%moElv_zgqfA4}Ckd%^{Z1~e=qVFgSkzk*1ca|3oUPhW{zZi%} zYt6Vwd-0=DTb0W8R9nQBC$6XSIie>VYLP*}1lZ0@<^=EnqM#MP7@1ftRjIuUcSB&J z&#Hyg1any^uyc}~CuN!s8=T=H|2jTLuyKkB=N)=IYR|3>mM+J8hG74Cphw zCkgm+?8@b>NQy;Bw*0*CSe63tkVrjsM z64!o{4wrj`8@#Sv$KKf3qwC)DJ2ZBk42=GZVjTqyT}fS)07cPXoi$=K%8{V~PJseC z6G;YU3Wq?EXiqeGkizlnl8Ct79|$Me6d5V7agCLj1P3R3_j=0bP%$|anUHCrYP>Hg z!N*3Hw7Tt4>_cUbJrqcXws7&fg+rBq&)2PHV0&90K%S?stLsUzm4YGBE~Fe<#L4 zSqQP?s4vraI-{62p1(Zub+~(`9Mij#&f=|$#NycWHT2myH+s1dU;RMG!0k-w{Jr&d zs_m2-b_bkAX=V#z6_wYw5r<;rN?Qh5dc{SKI%*Sc=g`oSiA93|?L?*&?SRe{42bvhPz|p z{&%aH?(A&2BF3zm2>U@NZ`Gd*H{Th{vq!$mj;$);GR&LnEwfsMHS(=CXasI-^vU!f z3l*NO6Y@k)ioTf26C{C*ojcheze(yeI8p2J50S}V-4(ULTl*|T(Oqdd+cra?5Vs@u zsYB&P6a@4U0#?3*TWqjbAY&Y9NHDr#DP;8Y2jgXXOL3cfB8s}uvUzq)#+sAY<74li zKqpp^!#!}fTB)}LE9dk4YAwC~PE$XH4Q4+%kB>@Xn{=}Dv6>|E8F2rnh4@0cut(uj zA|+E^YovKxCYBN&8suKA=#U;&O0g=1H9^p^r9)0qP~ODA$6i(p>P{jFEi7C6+S%MI z|2)MOahW;D7e1fYMy*2%UU8!1&68*N6Fx9+f6=yfsjbeM_z|!;uS9D2=3#Vyb8po{ zX~xyHAHD^6aE(};m~AOV94_Tm6aZ3gD1(72yX*o$cU|Jk4O$5C`39|LN@vaf&pBVW zc6kOvR)aS4Lh+z)kydkQ*4&H26}4IIf~6moa{ATtxoj4zCQJ91ziaI-4Qu9biaIWL zOoKPStQgl4yc{tQClKc?y-ZaXo-+1m`*d0E=V@SU>gfvJB@51mm$)bSHupqgE#QHN zoLo;=p9jZsPp#rD*XR3cEMMNpTsA8wS%`1SNpK7b(ON0?ykWqgjhw?SaNwU+U&>(+ zJSigRB;>`WG=%o4u9G6e1Cdu%d)8UdX$CVZ|Y?dA#r-@k_)l05ZIu{~7P$Q0?fIBu^@d`HkR9(41 zq_EaISKNDe28G-0JJK}dL`@}L-Nm8Url8apg{<~`I<6$BmMZn{icVb- z{IIWTO}DNTKW4eu^ux!$cDaph5;=;y$fcjVCK^3e(5T;6V_S4>D7MH`Q-ZJv^w3UK zmTofB_qm#Eb!?icb@#Tl21F84*E|8&1oDow^5Y;y`VvbEQB>}odCSKKQ{Er7w|B<@ z-;W)Ai51ZF8w?bv59Aq-vifSjCT>t8w7AY9Y?gm}Hv=6zdJ~M}6Tcey2#bDhj+Oci zv*4yPmJAaP%+j4FsuI%VQc3ILj_*)mxASSBgFN_d2n;bb^Q@wtsKw%!!wb>%im*bXjWwXGF#lxdt!XM8B zTA64Rz=esGRjq|CYj??%GcAU;b5~7znU@y^TyJR#7=L6($@HXzzy9Nn0XoM%(6C;; zAl-ahh1A1j@SkRd~oDBexp%UfUv=XV5}roNe9 zFyht-c0JLZ-0O|5y51&m1{e=|;Kxedqu50UMx2vSiSEb@PruGL!^0bVQxx{+!aCKg zS1)6WxOs^UFyER*rU6dsjgZx3Y?FIut>svg+idu8MEI2wkNT)qfxi74@_Zggvaw&E z`cHdpm#)H-r~XdZvW8$VO@|!V`)%<@v+)U#*VU;te)~9e^~**R+r^J{O0TF|)~X*} zSyS;|HGp#D)wjWWsXe60X~32=Yj{sjISzKsm_WG4OW{+wY!{EC;Wl5kUAE}&ilL^y zMeoi??d{&!?>^r}sJr!xFihLf(kym`y&}0_!)oC=7?b$|L;8;Q=NHi*+o77SgPCXs zvOCX0v3MM5OEg6IvI{p z7Y?v9*kyDX9Zt!+lYzz(H2Hp-9MAOomO0ZF_`T8SnII|IkWM9%d*(o=jyXC*K0kr_ zD|^9@XGUop$V>ssJLt;YzzPmN8^*6J!m-yt=zt|g60DTKm|f-(CD%oj-%x{vg{I;s zCHyz4f2zJK6%O`WY7Gb06oZYsFjy1yxtWCMyTFTs7EFa(FP+40&J&L4X4!Yj^)3VN zW^KB11k8LgPx`ZO4ZR+#NlbRa@$bRDKx9c-qLGzyH>S%yx2~wW&%MS3X6{oqUF(j} zM+OdTy3|ayz(QF~V}4=f;TxIk9}kXcR;}I%7*^x@=hj zwc#trKAd$gEX0@Euo4PE|4wRZ0Em8Y`Yl?+Q~;H|Iad^uL6bc#mU6V$fN*D%g10|T z$q&IaM864r#mAz*MnXnZVWtk-v=T=z+nVs&Y3$kKbv}pLz6;XYce?S_ul;ZE<=jE@qINGu{W8 z6SS#xclHQkcNxeK6}G+R#fkCfQ1_@Q;1&MRZr@IAmDde<_%mxjzB3qM7TypP*%0>h zLjBDxo(;?44S~&6K1=mh-em{5SV-IVFDqDf=`9wqEtXe-$hr93LS*V#?}}tzx(W^y zi4G?Jcpnt#6L5NjL$f~Uh4wkYZ0GXw%`f$Vhnui*Mnj*DbDXfGNE`G*_+WiI>|Fuq zt&?SeH!VT*^P}s)^q?#y$024%8^c|0mEZ2#N1$Z5)ljiC&u1@>fZmBRsy884I1^Ra zu6-8q8;o>D?)AMd;8PVSPG<48-Gvz`ACt8YTOT_`L;|mqbn|X!$r|gbaF(7ZKPT_G zN)uT?I|?u9O{W?H?N{Q02xfTEfAORDQL*prET75r{~aT_|1o2?1r>_$pPuNzfLq`( zR^G@}-m+gb2MF^{wf8`+$T>jt={}_J6=L`?!|>p`_ZwiJIPe#q<(DKWpHI+V!f|WR z7&YqA8cW3>jRX^lyE?941{7_as_PvTJE1IXRqV>Yh;5qbpU6d4A7=hd0|EXo5u~qe zW6P)Q=H|ulpASRYd`hmiZVdmal%AKeMq>Onwm^Pc3y3Aa%34SO0p_p@g!37{lwtmL0+%$l=O>v{ZBRr6b&m*8A>ffQe9m)wiRICm!d@!PQ^Bemd-Dm;^tB8vfrO;H{U1rJekyll2024-07-27T17:49:29-04:00https://zitniklab.hms.harvard.edu/feed.xmlZitnik LabHarvard Machine Learning for Medicine and ScienceMarinka ZitnikPINNACLE in Nature Methods2024-07-27T00:00:00-04:002024-07-27T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/07/27/PINNACLENatureMethods<p>PINNACLE contextual AI model is published in Nature Methods. <a href="https://www.nature.com/articles/s41592-024-02341-3">Paper.</a> <a href="https://www.nature.com/articles/s41592-024-02342-2">Research Briefing.</a> <a href="https://zitniklab.hms.harvard.edu/projects/PINNACLE/">Project website.</a></p>Marinka Zitnik

PINNACLE contextual AI model is published in Nature Methods. Paper. Research Briefing. Project website.Digital Twins as Global Health and Disease Models of Individuals2024-07-15T00:00:00-04:002024-07-15T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/07/15/DigitalTwins<p><a href="https://www.preprints.org/manuscript/202406.0357/">Paper on digitial twins</a> outlining strategies to leverage molecular and computational techniques to construct dynamic digital twins on the scale of populations to individuals.</p>Marinka ZitnikPaper on digitial twins outlining strategies to leverage molecular and computational techniques to construct dynamic digital twins on the scale of populations to individuals.Graph Diffusion Convolutions at ICML2024-07-14T00:00:00-04:002024-07-14T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/07/14/GraphAdversarialDiffusion<p><a href="https://arxiv.org/abs/2406.02059">Graph diffusion convolution is a geometric deep learning architecture that aggregates information from higher-order network neighbors through a generalized graph diffusion</a> to enhance model robustness to noisy and incomplete datasets. <a href="https://arxiv.org/abs/2406.02059">Paper at ICML.</a></p>Marinka ZitnikGraph diffusion convolution is a geometric deep learning architecture that aggregates information from higher-order network neighbors through a generalized graph diffusion to enhance model robustness to noisy and incomplete datasets. Paper at ICML.Three Papers: TrialBench, 3D Structure Design, LLM Editing2024-07-13T00:00:00-04:002024-07-13T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/07/13/TrialBenchLLM3DStructure<p>New papers on <a href="https://arxiv.org/abs/2407.00631">TrialBench with AI-ready clinical trial datasets</a>, <a href="https://arxiv.org/abs/2406.03403">structure-based drug design benchmark</a>, and <a href="https://arxiv.org/abs/2407.06483">composable interventions for language models</a>.</p>Marinka ZitnikNew papers on TrialBench with AI-ready clinical trial datasets, structure-based drug design benchmark, and composable interventions for language models.TDC-2: Multimodal Foundation for Therapeutics2024-06-23T00:00:00-04:002024-06-23T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/06/23/TDC2<p><a href="https://tdcommons.ai/">The Commons 2.0 (TDC-2)</a> is an overhaul of Therapeutic Data Commons to catalyze research in multimodal models for drug discovery by unifying single-cell biology of diseases, biochemistry of molecules, and effects of drugs through multimodal datasets, AI-powered API endpoints, new tasks and benchmarks. <a href="https://www.biorxiv.org/content/10.1101/2024.06.12.598655v2">Our paper.</a></p>Marinka ZitnikThe Commons 2.0 (TDC-2) is an overhaul of Therapeutic Data Commons to catalyze research in multimodal models for drug discovery by unifying single-cell biology of diseases, biochemistry of molecules, and effects of drugs through multimodal datasets, AI-powered API endpoints, new tasks and benchmarks. Our paper.Broad MIA: Protein Language Models2024-05-29T00:00:00-04:002024-05-29T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/05/29/BroadMIA<p>Check out our Broad’s seminars on <a href="https://youtu.be/LcLmvtXHI1s?si=GABqGdFt5ze9leT_">Multimodal protein language models for deciphering protein function.</a></p>Marinka ZitnikCheck out our Broad’s seminars on Multimodal protein language models for deciphering protein function.On Knowing a Gene in Cell Systems2024-05-28T00:00:00-04:002024-05-28T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/05/28/KnowingAGene<p>We shed light on <a href="https://www.sciencedirect.com/science/article/pii/S2405471224001236">distributional gene representations and their potential applications in biology to characterize gene function from a broader and more holistic perspective.</a></p>Marinka ZitnikWe shed light on distributional gene representations and their potential applications in biology to characterize gene function from a broader and more holistic perspective.Biomedical AI Agents2024-04-04T00:00:00-04:002024-04-04T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/04/04/BIomedicalAIAgents<p>We envision <a href="https://arxiv.org/abs/2404.02831">‘AI scientists’ as systems capable of skeptical learning and reasoning that empower biomedical research through collaborative agents</a> that integrate machine learning tools with experimental platforms.</p>Marinka ZitnikWe envision ‘AI scientists’ as systems capable of skeptical learning and reasoning that empower biomedical research through collaborative agents that integrate machine learning tools with experimental platforms.Efficient ML Seminar Series2024-03-23T00:00:00-04:002024-03-23T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/03/23/EfficientMLSeminar<p>We started a <a href="https://efficientml.org/">Harvard University Efficient ML Seminar Series</a>. Congrats to Jonathan for spearheading this initiative. <a href="https://www.harvardmagazine.com/2024/03/scaling-artificial-intelligence">Harvard Magazine</a> covered the first meeting focusing on LLMs.</p>Marinka ZitnikWe started a Harvard University Efficient ML Seminar Series. Congrats to Jonathan for spearheading this initiative. Harvard Magazine covered the first meeting focusing on LLMs.UniTS - Unified Time Series Model2024-03-04T00:00:00-05:002024-03-04T00:00:00-05:00https://zitniklab.hms.harvard.edu/2024/03/04/UniTS<p><a href="https://arxiv.org/abs/2403.00131">UniTS is a unified time series model</a> that can process classification, forecasting, anomaly detection and imputation tasks within a single model with no task-specific modules. UniTS has zero-shot, few-shot, and prompt learning capabilities. <a href="https://zitniklab.hms.harvard.edu/projects/UniTS/">Project website.</a></p>Marinka ZitnikUniTS is a unified time series model that can process classification, forecasting, anomaly detection and imputation tasks within a single model with no task-specific modules. UniTS has zero-shot, few-shot, and prompt learning capabilities. Project website. \ No newline at end of file +Jekyll2024-08-18T16:00:56-04:00https://zitniklab.hms.harvard.edu/feed.xmlZitnik LabHarvard Machine Learning for Medicine and ScienceMarinka ZitnikHow Proteins Behave in Context2024-08-15T00:00:00-04:002024-08-15T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/08/15/PINNACLENews<p><a href="https://hms.harvard.edu/news/new-ai-tool-captures-how-proteins-behave-context">Harvard Medicine News</a> on our new AI tool that captures how proteins behave in context. <a href="https://kempnerinstitute.harvard.edu/research/deeper-learning/context-matters-for-foundation-models-in-biology/">Kempner Institute</a> on how context matters for foundation models in biology.</p>Marinka ZitnikHarvard Medicine News on our new AI tool that captures how proteins behave in context. Kempner Institute on how context matters for foundation models in biology.PINNACLE in Nature Methods2024-07-27T00:00:00-04:002024-07-27T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/07/27/PINNACLENatureMethods<p>PINNACLE contextual AI model is published in Nature Methods. <a href="https://www.nature.com/articles/s41592-024-02341-3">Paper.</a> <a href="https://www.nature.com/articles/s41592-024-02342-2">Research Briefing.</a> <a href="https://zitniklab.hms.harvard.edu/projects/PINNACLE/">Project website.</a></p>Marinka ZitnikPINNACLE contextual AI model is published in Nature Methods. Paper. Research Briefing. Project website.Digital Twins as Global Health and Disease Models of Individuals2024-07-15T00:00:00-04:002024-07-15T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/07/15/DigitalTwins<p><a href="https://www.preprints.org/manuscript/202406.0357/">Paper on digitial twins</a> outlining strategies to leverage molecular and computational techniques to construct dynamic digital twins on the scale of populations to individuals.</p>Marinka ZitnikPaper on digitial twins outlining strategies to leverage molecular and computational techniques to construct dynamic digital twins on the scale of populations to individuals.Graph Diffusion Convolutions at ICML2024-07-14T00:00:00-04:002024-07-14T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/07/14/GraphAdversarialDiffusion<p><a href="https://arxiv.org/abs/2406.02059">Graph diffusion convolution is a geometric deep learning architecture that aggregates information from higher-order network neighbors through a generalized graph diffusion</a> to enhance model robustness to noisy and incomplete datasets. <a href="https://arxiv.org/abs/2406.02059">Paper at ICML.</a></p>Marinka ZitnikGraph diffusion convolution is a geometric deep learning architecture that aggregates information from higher-order network neighbors through a generalized graph diffusion to enhance model robustness to noisy and incomplete datasets. Paper at ICML.Three Papers: TrialBench, 3D Structure Design, LLM Editing2024-07-13T00:00:00-04:002024-07-13T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/07/13/TrialBenchLLM3DStructure<p>New papers on <a href="https://arxiv.org/abs/2407.00631">TrialBench with AI-ready clinical trial datasets</a>, <a href="https://arxiv.org/abs/2406.03403">structure-based drug design benchmark</a>, and <a href="https://arxiv.org/abs/2407.06483">composable interventions for language models</a>.</p>Marinka ZitnikNew papers on TrialBench with AI-ready clinical trial datasets, structure-based drug design benchmark, and composable interventions for language models.TDC-2: Multimodal Foundation for Therapeutics2024-06-23T00:00:00-04:002024-06-23T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/06/23/TDC2<p><a href="https://tdcommons.ai/">The Commons 2.0 (TDC-2)</a> is an overhaul of Therapeutic Data Commons to catalyze research in multimodal models for drug discovery by unifying single-cell biology of diseases, biochemistry of molecules, and effects of drugs through multimodal datasets, AI-powered API endpoints, new tasks and benchmarks. <a href="https://www.biorxiv.org/content/10.1101/2024.06.12.598655v2">Our paper.</a></p>Marinka ZitnikThe Commons 2.0 (TDC-2) is an overhaul of Therapeutic Data Commons to catalyze research in multimodal models for drug discovery by unifying single-cell biology of diseases, biochemistry of molecules, and effects of drugs through multimodal datasets, AI-powered API endpoints, new tasks and benchmarks. Our paper.Broad MIA: Protein Language Models2024-05-29T00:00:00-04:002024-05-29T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/05/29/BroadMIA<p>Check out our Broad’s seminars on <a href="https://youtu.be/LcLmvtXHI1s?si=GABqGdFt5ze9leT_">Multimodal protein language models for deciphering protein function.</a></p>Marinka ZitnikCheck out our Broad’s seminars on Multimodal protein language models for deciphering protein function.On Knowing a Gene in Cell Systems2024-05-28T00:00:00-04:002024-05-28T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/05/28/KnowingAGene<p>We shed light on <a href="https://www.sciencedirect.com/science/article/pii/S2405471224001236">distributional gene representations and their potential applications in biology to characterize gene function from a broader and more holistic perspective.</a></p>Marinka ZitnikWe shed light on distributional gene representations and their potential applications in biology to characterize gene function from a broader and more holistic perspective.Biomedical AI Agents2024-04-04T00:00:00-04:002024-04-04T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/04/04/BIomedicalAIAgents<p>We envision <a href="https://arxiv.org/abs/2404.02831">‘AI scientists’ as systems capable of skeptical learning and reasoning that empower biomedical research through collaborative agents</a> that integrate machine learning tools with experimental platforms.</p>Marinka ZitnikWe envision ‘AI scientists’ as systems capable of skeptical learning and reasoning that empower biomedical research through collaborative agents that integrate machine learning tools with experimental platforms.Efficient ML Seminar Series2024-03-23T00:00:00-04:002024-03-23T00:00:00-04:00https://zitniklab.hms.harvard.edu/2024/03/23/EfficientMLSeminar<p>We started a <a href="https://efficientml.org/">Harvard University Efficient ML Seminar Series</a>. Congrats to Jonathan for spearheading this initiative. <a href="https://www.harvardmagazine.com/2024/03/scaling-artificial-intelligence">Harvard Magazine</a> covered the first meeting focusing on LLMs.</p>Marinka ZitnikWe started a Harvard University Efficient ML Seminar Series. Congrats to Jonathan for spearheading this initiative. Harvard Magazine covered the first meeting focusing on LLMs. \ No newline at end of file diff --git a/index.html b/index.html index 4eb9d9a1..f353966f 100644 --- a/index.html +++ b/index.html @@ -177,6 +177,34 @@

AI for Science | Therapeutic Science

+
+
+ +
+ +

Aug 2024:   How Proteins Behave in Context

+ + +
+ +
+ + + +

Harvard Medicine News on our new AI tool that captures how proteins behave in context. Kempner Institute on how context matters for foundation models in biology.

+ + + + + +
+ + + +
+
+
@@ -721,36 +749,6 @@

AI for Science | Therapeutic Science

-
-
- -
- -

Oct 2023:   Structure-Based Drug Design

- - -
- -
- - - -

Geometric deep learning has emerged as a valuable tool for structure-based drug design, to generate and refine biomolecules by leveraging detailed three-dimensional geometric and molecular interaction information.

- - - - - - -
- - - -
-
-
diff --git a/jobs/index.html b/jobs/index.html index 9fd146ed..58ddce83 100644 --- a/jobs/index.html +++ b/jobs/index.html @@ -140,7 +140,7 @@
-

Thank you for being so interested in joining our group! Impactful research requires excellent mentoring. Prof. Zitnik is the recipient of the Young Mentor Award at Harvard Medical School—this prestigious award acknowledges that recognition to her.

+

Thank you for being so interested in joining our group! Impactful research requires excellent mentoring. Prof. Zitnik is the recipient of the Young Mentor Award at Harvard Medical School—this prestigious award acknowledges that recognition to her.

Graduate students

@@ -170,9 +170,9 @@

Postdoct

Postdoctoral research fellows in foundation AI

-

We have multiple openings for postdoctoral research fellows in the broad area of foundation models focusing on geometric deep learning, multimodal learning, large-scale knowledge graphs, large language models, generative AI, and AI agents.

+

We have multiple openings for postdoctoral research fellows in foundation models focusing on geometric deep learning, multimodal learning, large-scale knowledge graphs, large language models, generative AI, and AI agents.

-

Applications are reviewed on a rolling basis. Interested candidates are encouraged to submit their applications as soon as possible.

+

This position is available immediately. Interested candidates are encouraged to submit their applications as soon as possible.

NOW OPEN: Request For Applications

@@ -198,7 +198,7 @@

Postdoctoral research fe

Postdoctoral research fellows in medical AI

-

We have an opening for a postdoctoral research fellowship in novel methods in the broad area of medical AI.

+

We have multiple openings for postdoctoral research fellows in medical AI.

This position is available immediately. Interested candidates are encouraged to submit their applications as soon as possible.

@@ -259,6 +259,34 @@

Visitors, interns, and short-t
+
+
+ +
+ +

Aug 2024:   How Proteins Behave in Context

+ + +
+ +
+ + + +

Harvard Medicine News on our new AI tool that captures how proteins behave in context. Kempner Institute on how context matters for foundation models in biology.

+ + + + + +
+ + + +
+
+
@@ -803,36 +831,6 @@

Visitors, interns, and short-t

-
-
- -
- -

Oct 2023:   Structure-Based Drug Design

- - -
- -
- - - -

Geometric deep learning has emerged as a valuable tool for structure-based drug design, to generate and refine biomolecules by leveraging detailed three-dimensional geometric and molecular interaction information.

- - - - - - -
- - - -
-
-
diff --git a/meetings/index.html b/meetings/index.html index d27a6356..3c3bba94 100644 --- a/meetings/index.html +++ b/meetings/index.html @@ -594,6 +594,34 @@

Biomedical Data Fusion (EMBC a
+
+
+ +
+ +

Aug 2024:   How Proteins Behave in Context

+ + +
+ +
+ + + +

Harvard Medicine News on our new AI tool that captures how proteins behave in context. Kempner Institute on how context matters for foundation models in biology.

+ + + + + +
+ + + +
+
+
@@ -1138,36 +1166,6 @@

Biomedical Data Fusion (EMBC a

-
-
- -
- -

Oct 2023:   Structure-Based Drug Design

- - -
- -
- - - -

Geometric deep learning has emerged as a valuable tool for structure-based drug design, to generate and refine biomolecules by leveraging detailed three-dimensional geometric and molecular interaction information.

- - - - - - -
- - - -
-
-
diff --git a/news/index.html b/news/index.html index b65ffa90..ea9acd03 100644 --- a/news/index.html +++ b/news/index.html @@ -170,6 +170,34 @@
+
+
+ +
+ +

Aug 2024:   How Proteins Behave in Context

+ + +
+ +
+ + + +

Harvard Medicine News on our new AI tool that captures how proteins behave in context. Kempner Institute on how context matters for foundation models in biology.

+ + + + + +
+ + + +
+
+
@@ -1594,34 +1622,6 @@
-
-
- -
- -

Aug 2022:   On Graph AI for Precision Medicine

- - -
- -
- - - -

The recording of our tutorial on using graph AI to advance precision medicine is available. Tune into four hours of interactive lectures about state-of-the-art graph AI methods and applications in precision medicine.

- - - - - -
- - - -
-
-
+
+
+ +
+ +

Aug 2022:   On Graph AI for Precision Medicine

+ + +
+ +
+ + + +

The recording of our tutorial on using graph AI to advance precision medicine is available. Tune into four hours of interactive lectures about state-of-the-art graph AI methods and applications in precision medicine.

+ + + + + +
+ + + +
+
+
@@ -1553,34 +1581,6 @@
-
-
- -
- -

Mar 2021:   Hot Off the Press: Multiscale Interactome

- - -
- -
- - - -

Hot off the press! We develop a multiscale interactome approach to explain disease treatments. The approach can predict drug-disease treatments, identify proteins and biological functions related to treatment, and identify genes that alter treatment’s efficacy and adverse reactions.

- - - - - -
- - - -
-
-
+
+
+ +
+ +

Mar 2021:   Hot Off the Press: Multiscale Interactome

+ + +
+ +
+ + + +

Hot off the press! We develop a multiscale interactome approach to explain disease treatments. The approach can predict drug-disease treatments, identify proteins and biological functions related to treatment, and identify genes that alter treatment’s efficacy and adverse reactions.

+ + + + + +
+ + + +
+
+
diff --git a/people/index.html b/people/index.html index ebda32f1..3019cd92 100644 --- a/people/index.html +++ b/people/index.html @@ -674,20 +674,20 @@
- +
- <a href=Pengwei Sui" /> + <a href=Michelle Dai" />
-

Pengwei Sui

+

Michelle Dai

Research Associate

@@ -703,20 +703,20 @@
- +
- <a href=Owen Queen" /> + <a href=Pengwei Sui" />
-

Owen Queen

+

Pengwei Sui

Research Associate

@@ -732,21 +732,21 @@
- +
- <a href=Zaixi Zhang" /> + <a href=Owen Queen" />
-

Zaixi Zhang

-

Visiting PhD Student

+

Owen Queen

+

Research Associate

@@ -761,21 +761,21 @@
- +
- <a href=Michelle Dai" /> + <a href=Zaixi Zhang" />
-

Michelle Dai

-

Graduate Researcher

+

Zaixi Zhang

+

Visiting PhD Student

@@ -1012,6 +1012,34 @@

Lab alumni

+
+
+ +
+ +

Aug 2024:   How Proteins Behave in Context

+ + +
+ +
+ + + +

Harvard Medicine News on our new AI tool that captures how proteins behave in context. Kempner Institute on how context matters for foundation models in biology.

+ + + + + +
+ + + +
+
+
@@ -1556,36 +1584,6 @@

Lab alumni

-
-
- -
- -

Oct 2023:   Structure-Based Drug Design

- - -
- -
- - - -

Geometric deep learning has emerged as a valuable tool for structure-based drug design, to generate and refine biomolecules by leveraging detailed three-dimensional geometric and molecular interaction information.

- - - - - - -
- - - -
-
-
diff --git a/postdoc-ML/index.html b/postdoc-ML/index.html index ee3999c3..0b489a84 100644 --- a/postdoc-ML/index.html +++ b/postdoc-ML/index.html @@ -154,7 +154,7 @@

Qualifications

We are looking for applicants with demonstrably strong research skills, ideally, with multiple publications in top venues in machine learning and artificial intelligence, and/or top-tier scientific journals.

-

Candidates must have a Ph.D. or equivalent degree in computer science, statistics, or a closely related field. Strong programming skills and practical experience with leading machine learning frameworks are required. Experience and/or interest in applications of AI to science, biology and medicine is a strong plus.

+

Candidates must have a Ph.D. or equivalent degree in computer science. Strong programming skills and practical experience with leading machine learning frameworks are required. Experience and/or interest in applications of AI to science, biology and medicine is a strong plus.

Application process

@@ -191,6 +191,34 @@

Advisor

+
+
+ +
+ +

Aug 2024:   How Proteins Behave in Context

+ + +
+ +
+ + + +

Harvard Medicine News on our new AI tool that captures how proteins behave in context. Kempner Institute on how context matters for foundation models in biology.

+ + + + + +
+ + + +
+
+
@@ -735,36 +763,6 @@

Advisor

-
-
- -
- -

Oct 2023:   Structure-Based Drug Design

- - -
- -
- - - -

Geometric deep learning has emerged as a valuable tool for structure-based drug design, to generate and refine biomolecules by leveraging detailed three-dimensional geometric and molecular interaction information.

- - - - - - -
- - - -
-
-
diff --git a/postdoc-TDC/index.html b/postdoc-TDC/index.html index 26741e11..1403d8cf 100644 --- a/postdoc-TDC/index.html +++ b/postdoc-TDC/index.html @@ -203,6 +203,34 @@

Advisor

+
+
+ +
+ +

Aug 2024:   How Proteins Behave in Context

+ + +
+ +
+ + + +

Harvard Medicine News on our new AI tool that captures how proteins behave in context. Kempner Institute on how context matters for foundation models in biology.

+ + + + + +
+ + + +
+
+
@@ -747,36 +775,6 @@

Advisor

-
-
- -
- -

Oct 2023:   Structure-Based Drug Design

- - -
- -
- - - -

Geometric deep learning has emerged as a valuable tool for structure-based drug design, to generate and refine biomolecules by leveraging detailed three-dimensional geometric and molecular interaction information.

- - - - - - -
- - - -
-
-
diff --git a/postdoc-biomedicalAI-MGB/index.html b/postdoc-biomedicalAI-MGB/index.html index 3b557a4b..75d15d0f 100644 --- a/postdoc-biomedicalAI-MGB/index.html +++ b/postdoc-biomedicalAI-MGB/index.html @@ -180,6 +180,34 @@

Application process

+
+
+ +
+ +

Aug 2024:   How Proteins Behave in Context

+ + +
+ +
+ + + +

Harvard Medicine News on our new AI tool that captures how proteins behave in context. Kempner Institute on how context matters for foundation models in biology.

+ + + + + +
+ + + +
+
+
@@ -724,36 +752,6 @@

Application process

-
-
- -
- -

Oct 2023:   Structure-Based Drug Design

- - -
- -
- - - -

Geometric deep learning has emerged as a valuable tool for structure-based drug design, to generate and refine biomolecules by leveraging detailed three-dimensional geometric and molecular interaction information.

- - - - - - -
- - - -
-
-
diff --git a/postdoc-cancerTxAI/index.html b/postdoc-cancerTxAI/index.html index 72e971e8..c87ea0b3 100644 --- a/postdoc-cancerTxAI/index.html +++ b/postdoc-cancerTxAI/index.html @@ -189,6 +189,34 @@

Faculty and mentors

+
+
+ +
+ +

Aug 2024:   How Proteins Behave in Context

+ + +
+ +
+ + + +

Harvard Medicine News on our new AI tool that captures how proteins behave in context. Kempner Institute on how context matters for foundation models in biology.

+ + + + + +
+ + + +
+
+
@@ -733,36 +761,6 @@

Faculty and mentors

-
-
- -
- -

Oct 2023:   Structure-Based Drug Design

- - -
- -
- - - -

Geometric deep learning has emerged as a valuable tool for structure-based drug design, to generate and refine biomolecules by leveraging detailed three-dimensional geometric and molecular interaction information.

- - - - - - -
- - - -
-
-
diff --git a/postdoc-medicalAI/index.html b/postdoc-medicalAI/index.html index 0dfd3f92..253116e1 100644 --- a/postdoc-medicalAI/index.html +++ b/postdoc-medicalAI/index.html @@ -159,7 +159,7 @@

Qualifications

We are looking for applicants with demonstrably strong research skills, ideally, with multiple publications in top venues in machine learning and artificial intelligence, and/or top-tier medical journals.

-

Candidates must have a Ph.D. or equivalent degree in computer science, statistics, or a closely related field. Strong programming skills and practical experience with leading machine learning frameworks are required. Experience and/or interest in medical AI is a strong plus.

+

Candidates must have a Ph.D. or equivalent degree in computer science. Strong programming skills and practical experience with leading machine learning frameworks are required. Experience and/or interest in medical AI is a strong plus.

Application process

@@ -196,6 +196,34 @@

Advisor

+
+
+ +
+ +

Aug 2024:   How Proteins Behave in Context

+ + +
+ +
+ + + +

Harvard Medicine News on our new AI tool that captures how proteins behave in context. Kempner Institute on how context matters for foundation models in biology.

+ + + + + +
+ + + +
+
+
@@ -740,36 +768,6 @@

Advisor

-
-
- -
- -

Oct 2023:   Structure-Based Drug Design

- - -
- -
- - - -

Geometric deep learning has emerged as a valuable tool for structure-based drug design, to generate and refine biomolecules by leveraging detailed three-dimensional geometric and molecular interaction information.

- - - - - - -
- - - -
-
-
diff --git a/products/ada_fang/index.html b/products/ada_fang/index.html index 0822d1a5..4e6f1239 100644 --- a/products/ada_fang/index.html +++ b/products/ada_fang/index.html @@ -23,14 +23,14 @@ - + +{"image":"https://zitniklab.hms.harvard.edu/img/ada_fang.png","url":"https://zitniklab.hms.harvard.edu/products/ada_fang/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/ada_fang/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Ada Fang","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/alex_verce.png","url":"https://zitniklab.hms.harvard.edu/products/alejandro_velez_arce/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/alejandro_velez_arce/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Alejandro Velez Arce","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/andrew_shen.png","url":"https://zitniklab.hms.harvard.edu/products/andrew_shen/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/andrew_shen/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Andrew Shen","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/ayush_noori.png","url":"https://zitniklab.hms.harvard.edu/products/ayush_noori/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/ayush_noori/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Ayush Noori","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/george_dasoulas.png","url":"https://zitniklab.hms.harvard.edu/products/george_dasoulas/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/george_dasoulas/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"George Dasoulas","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/intae_moon.png","url":"https://zitniklab.hms.harvard.edu/products/intae_moon/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/intae_moon/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Intae Moon","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/jason_poulos.png","url":"https://zitniklab.hms.harvard.edu/products/jason_poulos/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/jason_poulos/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Jason Poulos","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} - -Jonathan Richard Schwarz | Zitnik Lab - - - - - - - - - - - - - - - - - - - - - - - - - - - -
-
-
-

Jonathan Richard Schwarz

-

Postdoctoral Fellow

- -
-
-
- - - - -
-
-
- -
- - -
- -
-
- -
-
- -
-

Jonathan Richard Schwarz

-

Postdoctoral Fellow

-

- - - - - -
- -
-

Description

-
- - - - -
-
- - - -
- - -
- -
-
-
- - - - - diff --git a/products/kevin_li/index.html b/products/kevin_li/index.html index 8b678243..642e1700 100644 --- a/products/kevin_li/index.html +++ b/products/kevin_li/index.html @@ -23,14 +23,14 @@ - + +{"image":"https://zitniklab.hms.harvard.edu/img/kevin_li.png","url":"https://zitniklab.hms.harvard.edu/products/kevin_li/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/kevin_li/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Kevin Li","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} - -Kushan Weerakoon | Zitnik Lab - - - - - - - - - - - - - - - - - - - - - - - - - - - -
-
-
-

Kushan Weerakoon

-

Undergraduate Researcher
Harvard

- -
-
-
- - - - -
-
-
- -
- - -
- -
-
- -
-
- -
-

Kushan Weerakoon

-

Undergraduate Researcher
Harvard

-

- - - - - -
- -
-

Description

-
- - - - -
-
- - - -
- - -
- -
-
-
- - - - - diff --git a/products/marinka_zitnik/index.html b/products/marinka_zitnik/index.html index 1e8a74b1..59fe8edb 100644 --- a/products/marinka_zitnik/index.html +++ b/products/marinka_zitnik/index.html @@ -23,14 +23,14 @@ - + +{"image":"https://zitniklab.hms.harvard.edu/img/marinka_zitnik.png","url":"https://zitniklab.hms.harvard.edu/products/marinka_zitnik/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/marinka_zitnik/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Marinka Zitnik","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/michelle_dai.png","url":"https://zitniklab.hms.harvard.edu/products/michelle_dai/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/michelle_dai/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Michelle Dai","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/michelle_li.png","url":"https://zitniklab.hms.harvard.edu/products/michelle_li/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/michelle_li/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Michelle M. Li","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/owen_queen.png","url":"https://zitniklab.hms.harvard.edu/products/owen_queen/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/owen_queen/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Owen Queen","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/pengwei_sui.png","url":"https://zitniklab.hms.harvard.edu/products/pengwei_sui/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/pengwei_sui/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Pengwei Sui","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/richard_zhu.png","url":"https://zitniklab.hms.harvard.edu/products/richard_zhu/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/richard_zhu/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Richard Zhu","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/robert_calef.png","url":"https://zitniklab.hms.harvard.edu/products/robert_calef/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/robert_calef/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Robert Calef","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/ruth_johnson.png","url":"https://zitniklab.hms.harvard.edu/products/ruth_johnson/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/ruth_johnson/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Ruth Johnson","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/shanghua_gao.png","url":"https://zitniklab.hms.harvard.edu/products/shanghua_gao/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/shanghua_gao/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Shanghua Gao","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/valentina_giunchiglia.png","url":"https://zitniklab.hms.harvard.edu/products/valentina_giunchiglia/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/valentina_giunchiglia/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Valentina Giunchiglia","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/wanxiang_shen.png","url":"https://zitniklab.hms.harvard.edu/products/wanxiang_shen/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/wanxiang_shen/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Wanxiang Shen","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/xiang_lin.png","url":"https://zitniklab.hms.harvard.edu/products/xiang_lin/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/xiang_lin/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Xiang Lin","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/xiaorui_su.png","url":"https://zitniklab.hms.harvard.edu/products/xiaorui_su/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/xiaorui_su/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Xiaorui Su","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/yasha_ektefaie.png","url":"https://zitniklab.hms.harvard.edu/products/yasha_ektefaie/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/yasha_ektefaie/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Yasha Ektefaie","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Biomedical Machine Learning, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/yepeng_huang.png","url":"https://zitniklab.hms.harvard.edu/products/yepeng_huang/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/yepeng_huang/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Yepeng Huang","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/zaixi_zhang.png","url":"https://zitniklab.hms.harvard.edu/products/zaixi_zhang/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/zaixi_zhang/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Zaixi Zhang","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} +{"image":"https://zitniklab.hms.harvard.edu/img/zhenglun_kong.png","url":"https://zitniklab.hms.harvard.edu/products/zhenglun_kong/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/zhenglun_kong/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Zhenglun Kong","dateModified":"2024-08-18T16:00:56-04:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-08-18T16:00:56-04:00","@type":"BlogPosting","@context":"https://schema.org"} + + + + +
+
+
+
+ + + +
diff --git a/sitemap.xml b/sitemap.xml index 0b47d371..f30316de 100644 --- a/sitemap.xml +++ b/sitemap.xml @@ -545,104 +545,108 @@ 2024-07-27T00:00:00-04:00 +https://zitniklab.hms.harvard.edu/2024/08/15/PINNACLENews/ +2024-08-15T00:00:00-04:00 + + https://zitniklab.hms.harvard.edu/products/ada_fang/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/alejandro_velez_arce/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/andrew_shen/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/ayush_noori/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/george_dasoulas/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/intae_moon/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/jason_poulos/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/kevin_li/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/marinka_zitnik/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/michelle_dai/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/michelle_li/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/owen_queen/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/pengwei_sui/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/richard_zhu/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/robert_calef/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/ruth_johnson/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/shanghua_gao/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/valentina_giunchiglia/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/wanxiang_shen/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/xiang_lin/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/xiaorui_su/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/yasha_ektefaie/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/yepeng_huang/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/zaixi_zhang/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/products/zhenglun_kong/ -2024-07-27T17:49:29-04:00 +2024-08-18T16:00:56-04:00 https://zitniklab.hms.harvard.edu/404/ @@ -786,10 +790,6 @@ https://zitniklab.hms.harvard.edu/news/page3/ -https://zitniklab.hms.harvard.edu/docs/marinka_zitnik_CV.pdf -2022-02-19T16:11:26-05:00 - - https://zitniklab.hms.harvard.edu/docs/slides/IJCAI-T27-000-intro.pdf 2021-01-07T14:36:27-05:00 diff --git a/software/index.html b/software/index.html index ec87bd98..524b8b12 100644 --- a/software/index.html +++ b/software/index.html @@ -1743,6 +1743,34 @@
+
+
+ +
+ +

Aug 2024:   How Proteins Behave in Context

+ + +
+ +
+ + + +

Harvard Medicine News on our new AI tool that captures how proteins behave in context. Kempner Institute on how context matters for foundation models in biology.

+ + + + + +
+ + + +
+
+
@@ -2287,36 +2315,6 @@
-
-
- -
- -

Oct 2023:   Structure-Based Drug Design

- - -
- -
- - - -

Geometric deep learning has emerged as a valuable tool for structure-based drug design, to generate and refine biomolecules by leveraging detailed three-dimensional geometric and molecular interaction information.

- - - - - - -
- - - -
-
-
diff --git a/talks/index.html b/talks/index.html index 52fd1ce3..612803f4 100644 --- a/talks/index.html +++ b/talks/index.html @@ -171,6 +171,11 @@