8_create_annotated_metatranscriptome.txt

#warming_AK

#Adriana L. Romero-Olivares' protocol for processing RNAseq data of soil 
#Section 8 
#File name: 8_create_annotated_metatranscriptome
#Step: Create annotated meta-transcriptome

#This protocol creates an annotated meta-transcriptome 
#Once it's created, follow steps 4, 5, and 6 for a list of up- and down-regulated genes
#Follow instructions from Section 7 to create an index for your database and hmmscan for longestORFs 
#Use .domtblout file 

#First, extract column with hits using awk 

awk '{print $4}' pfam_hmmscan_output.domtblout > pfam_hits_list 

#modify file to remove extra lines or header of file using sed
#this will be done by filtering out everything that doesn't contain the word TRINITY (in my case, all contigs have the word TRINITY, and that's all I want to keep) 

sed -i '/TRINITY/!d' hits2 #delete all lines that don't have the word TRINITY

#remove any additional info extracted with the contigs' name 

sed 's/|.*//' cazy_hits_list > cazy_hits_list_2

#weird step, but, rename cazy_hits_list_2 to cazy_hits_list

#now delete repeated lines so that you have a single hit for each protein  

awk '!seen[$0]++' cazy_hits_list > cazy_hits_list_clean

#generate file with fasta sequences from your meta-transcriptome 
#this will generate a fasta file that will only have annotated contigs = an annotated meta-transcriptome  

grep -v "^#" cazy_hits_list_clean | awk '{print $1}' | esl-sfetch -f Trinity_ALL_f.Trinity.fasta - > cazy_myhits.fa

#proceed to step 4 using cazy_myhits.fa as your meta-transcriptome 
#once you have the final list of genes up- and down-regulated you can see what those are using the domtblout table you used at the beginning of this protocol.