Nick's updates to 7.0. Updates to Makefile and README to fix issues.

.gitignore

@@ -13,34 +13,6 @@ src/admixlib/*
 src/admixtables/*
 
 #Various output files
-src/backup
-src/convertf
-src/copych
-src/diff.out
-src/diffch
-src/dirlist
-src/dowtjack
-src/gmon.out
-src/grabpars
-src/libnick.a
-src/nicksrc/tmp
-src/oldmake
-src/oldmakefile
-src/q1
-src/qaz
-src/qp3Pop
-src/qp4diff
-src/qpAdm
-src/qpBound
-src/qpDpart
-src/qpDstat
-src/qpff3base
-src/qpF4ratio
-src/qpGraph
-src/qpWave
-src/qpreroot
-src/rolloff
-src/tables
 pdoc.aux
 pdoc.dvi
 pdoc.log

README

@@ -1,4 +1,4 @@
-ADMIXTOOLS version 6.0  1/10/20 (for Linux and Mac)
+ADMIXTOOLS version 7.0  6/23/20 (for Linux and Mac)
 
 See README.INSTALL for installation info. 
 
@@ -26,9 +26,22 @@ recompile the programs, type
 
 If you are building on a Mac, you will need gsl and openblas installed. 
 "brew install gsl"
-"brew install homebrew/science/openblas"
+"brew install openblas"
 Uncomment the lines in src/Makefile that modify the CFLAGS and LDFLAGS. 
 
+If these packages are not in standard directories, you can specify the 
+appropriate include and library directories with the CFLAGS and LDFLAGS 
+make variables. 
+For example, on the Harvard Medical School O2 cluster, the command is:
+'make CFLAGS="-I/n/app/openblas/0.2.19/include -I/n/app/gsl/2.3/include" LDFLAGS="-L/n/app/openblas/0.2.19/lib -L/n/app/gsl/2.3/lib/"'
+On Mac OSX:
+'make CFLAGS="-I/usr/local/opt/openblas/include -I/usr/local/opt/gsl/include" LDFLAGS="-L/usr/local/opt/openblas/lib -L/usr/local/opt/gsl/lib"'
+
+If you have issues with missing lapacke symbols, for example "undefined reference to `LAPACKE_dlange'", run make with the corresponding additional libraries linked:
+'make LDLIBS="-llapacke"' on Linux Mint 18. 
+Similarly for Ubuntu, for "undefined reference to dspev_"
+'make LDLIBS="-llapack"'
+
 If you have trouble compiling and running our code, try compiling and
 running the examples in the examples/ directory. Download the example
 datasets from https://reich.hms.harvard.edu/sites/reich.hms.harvard.edu/files/inline-files/AdmixTools_Example_Data.tar_.gz

README.QPGRAPH

@@ -73,6 +73,16 @@ qpGraph will accept either style.
 [-d dotfile]  Graph output in "dot format";  dot -Tps < dotfile > dotfile.ps gives postscript
 "dot" must be installed (part of Graphviz) 
 
+*** new feature ***  
+Input can now be an fstats file -- see README.qpfstats
+*** new feature *** 
+halfscore (won't work in combination with fstats)
+halfscore:  YES 
+halfjackname:  <output file> 
+Does a VERY conservative goodness of fit test.  See halfscore.pdf for details.
+
+
+
 ===================================================================================================
 Utility program: qpreroot
 If the input graph is not connected or has loops qpGraph will die unpleasantly.  

README.QpWave

@@ -84,6 +84,10 @@ program code:
 
 diagplus:  0  
 
+*** new feature (for both qpAdm and qpWave)***  
+Input can now be an fstats file -- see README.qpfstats
+For qpAdm new parameter numboot: see README.qpfstats
+
 
 Nick Patterson
 <[email protected]>

README.qpfstats

@@ -0,0 +1,64 @@
+qpfstats
+--------
+         DIR:                 /home/np29/biologyx/v41x/yamdir2/testdir 
+          S1:                     qdata1  
+         S1X:                     qdata1  
+   indivname:                DIR/S1X.ind
+     snpname:                 DIR/S1.snp
+genotypename:                DIR/S1.geno
+ poplistname:                      lista
+## must be present.  ne popuation / line.  First population is base
+ fstatsoutname:                    fstatsa.txt 
+## first line is header.  Must be retained
+ allsnps:                          YES 
+## default NO  -- dangerous bend
+ inbreed:                           NO 
+## default NO
+ scale:                             NO 
+## default YES -- when fstats are scaled to "match" fst in least squares sense 
+
+qpfmv 
+-----
+ fstatsname:     fstatsa.txt
+ popfilename:      f4sslist
+## 4 pops / line (as in qpDstat) but f2, f3, f4 can be mixed.  So code A C B C for f3 stat 
+     fmvoutname:  fmvq2.out 
+
+
+New parameters for qpAdm (which should be upwards compatible) 
+fstatsname:   <output from qpfstats> 
+## if present no need to specify .ind .snp .geno 
+numboot:    <# of bootstrap samples + antithetical samples> 
+## default 1000 
+
+New parameters for qpWave (which should be upwards compatible) 
+fstatsname:   <output from qpfstats> 
+
+New parameters for qpGraph (which should be upwards compatible) 
+qpGraph allows to to make qpfstats file 
+Example 
+1) Make qpfstats  :: ~np29/newadm19/src/jtest1/ww2dir/testdir/qpfs_example/parw2 
+D2:          /home/np29/broaddatax/v41  
+D1:          D2
+S1:           jd1               
+indivname:    D1/S1.ind
+snpname:      D1/S1.snp 
+genotypename: D1/S1.geno
+fstatsoutname:  fstatsw2    
+allsnps:        YES
+## there is an option oldallsnps: YES for compatiblity.  Not recommended. 
+loadf3:         YES 
+## this is needed to get qpfstats calculated
+graphname:       graph1   
+diag:           .0001
+
+2) Use qpfstats  
+fstatsname:  fstatsw2    
+graphname:       graph1   
+diag:           .0001
+
+*** qpfstats with allsnps: YES  is a much better way to use all the data than 
+the older allsnps mode.   Standard errors are often much reduced and the theory 
+is now defensible!
+
+See ./qpfs.pdf for a brief explanation of the theory. 

fancy.txt

@@ -0,0 +1,17 @@
+Consider f4(l1, l2, r1, r2)  
+Suppose at some snp l2 is missing, but the allele frequencies of r1, r2 match.  Then whatever l2 f4 is zero  
+Similarly f4=0 if the frequncy of l1 and l2 match, 
+
+qpAdm
+qpWave
+qpF4ratio
+qpf4diff  
+
+use this trick. 
+
+Arie Shaus pointed out to me that this can bias f4 down toward 0.  
+To disable, code in the parameter file 
+
+fancyf4:  NO
+
+Nick  1/29/19 

perlsrc/jackdiff

@@ -0,0 +1,46 @@
+#!/usr/bin/perl  -w 
+
+use Getopt::Std ;
+getopts('a:b:o:',\%opts) ;
+
+$fa = $opts{"a"} ; 
+$fb = $opts{"b"} ; 
+$fout = $opts{"o"} ; 
+
+die "no -a parameter\n" unless (defined $fa) ;
+die "no -b parameter\n" unless (defined $fb) ;
+
+open (I1, "$fa") || die "can't open $fa\n" ; 
+open (I2, "$fb") || die "can't open $fa\n" ; 
+
+$maxa = -1 ; 
+foreach $line (<I1>) { 
+ ($a, $w, $v) = split " ", $line ; 
+ next if ($a =~ /\#/) ;
+ next unless (defined $v) ; 
+ $WA[$a] = $w ; 
+ $VA[$a] = $v ; 
+}
+foreach $line (<I2>) { 
+ ($a, $w, $v) = split " ", $line ; 
+ next if ($a =~ /\#/) ;
+ next unless (defined $v) ; 
+ $WB[$a] = $w ; 
+ $VB[$a] = $v ; 
+ $maxa = $a ; 
+}
+$T1 = "/tmp/t1.$$" ;
+open (Y1, ">$T1") ; 
+
+foreach $a (0..$maxa) { 
+ $wa = $WA[$a] ;
+ $wb = $WB[$a] ;
+ $va = $VA[$a] ;
+ $vb = $VB[$a] ;
+ die "mismatched weights: line index $a\n" if ($wa != $wb) ; 
+ printf Y1 "%d %d %12.6f\n", $a, $wa, $va - $vb ; 
+}
+close Y1 ; 
+system "mkpretty $T1 $fout" if (defined $fout) ; 
+printf "##jackknife difference $fa - $fb\n" ;
+system "simpjack2 -i $T1 -v" ; 

perlsrc/mkpretty

@@ -0,0 +1,66 @@
+#!/usr/bin/perl  -w 
+
+## mkpretty infile outfile 
+
+($infile, $outfile) = @ARGV ;
+$T1 = "/tmp/t1.$$" ;
+$T2 = "/tmp/t2.$$" ;
+
+if (defined $infile) {
+die "Usage: mkpretty < infile > outfile OR mkpretty infile outfile\n" unless (defined $outfile) ;
+system "cp $infile $T1" ;
+}
+else {
+system "cat - > $T1" ;
+}
+open (F1, $T1) || die "badly\n" ;
+for (;;) {
+ last if (eof(F1)) ;
+ $line = <F1> ;
+ chomp $line ;
+ (@Z) = split " ", $line ;
+ next unless (defined $Z[0]) ;
+ next if  ($Z[0] =~ /\#/) ;
+ $n = @Z ;
+ foreach $x (0..$n-1) { 
+  $W[$x] = 0 unless (defined ($W[$x])) ;
+  $w = length($Z[$x]) ;
+  $W[$x] = $w if ($w > $W[$x]) ;
+ }
+}
+## @W is maximum width
+close F1 ; 
+$nw = @W ; 
+foreach $x (0..$nw-1) { 
+ $w = $W[$x] ;
+ $fmt = "%" . $w . "s" ; 
+ $F[$x] = $fmt ;
+}
+
+open (F1, $T1) || die "badly\n" ;
+open (YY, ">$T2") || die "very badly\n" ;
+for (;;) {
+ last if (eof(F1)) ;
+ $line = <F1> ;
+ chomp $line ;
+ (@Z) = split " ", $line ;
+ next unless (defined $Z[0]) ;
+ if  ($Z[0] =~ /\#/) {  
+  print YY "$line\n";
+  next ;
+ }
+ $n = @Z ;
+ foreach $x (0..$n-1) { 
+  printf YY "$F[$x]", $Z[$x] ;
+  print  YY "  ", if ($x != ($n-1)) ;
+ }
+ print YY "\n" ;
+}
+if (defined $outfile) { 
+ system "cp $T2 $outfile" ;
+} 
+else { 
+ system "cat $T2" ;
+}
+unlink ($T1, $T2) ;
+

src/.gitignore

@@ -1 +1,37 @@
+backup
+convertf
+copych
+diff.out
+diffch
+dirlist
+dowtjack
 expfit.sh
+gcount
+gmon.out
+grabpars
+kimf
+libnick.a
+mergeit
+nicktmp
+oldmake
+oldmakefile
+q1
+qaz
+qp3Pop
+qp4diff
+qpAdm
+qpBound
+qpDpart
+qpdslow
+qpDstat
+qpff3base
+qpfmv
+qpfstats
+qpF4ratio
+qpGraph
+qpWave
+qpreroot
+rolloff
+rolloffp
+simpjack2
+tables