Salmon loop on multiple samples

Question

[m.storti]

Hi everyone!

I'm entering the bioinformatics world, and I have some problems with Salmon. In particular, I have to perform salmon on 101 pair-end samples (no replicates). I could do it one by one, but I think it is prone to errors. I tried some loops that I found on the net, but none of them worked. I have the index already. Please find below some of the samples that I have to align.

R55_S1_R2_001.fastq
R56_S2_R2_001.fastq
R57_S3_R2_001.fastq
R58_S4_R2_001.fastq
R59_S5_R2_001.fastq
R60_S6_R2_001.fastq
R61_S7_R2_001.fastq
R62_S8_R2_001.fastq

Do you have any suggestions? Many thanks in advance!

[Pierre Lindenbaum]

I tried some loops that I found on the net, but none of them worked.

show us what you tried

[m.storti]

Many thanks Pierre for your answer! I tried this:

#!/bin/bash

mkdir -p out/ ;
  mkdir -p out/salmon/ ;

 paste R1.list R2.list | while read R1 R2 ;
  do
echo -e "\n" ;
echo -e "/home/mstorti/FASTQ_Rabbit/run1_2/""${R1}"",""${R2}" ;
outdir=$(echo "${R2}" | cut -f7 -d"/" | cut -f1 -d"_") ;
mkdir -p "out/salmon/""${outdir}" ;

salmon quant \
  --index "/home/mstorti/FASTQ_Rabbit/run1_2/salmon/index" \
  --l A \
  -1 "${R1}" \
  -2 "${R2}" \
  --validateMappings \
  --output="out/salmon/""${outdir}" ;

echo "--Done." ;
  done

And this one:

#!/bin/bash


for fn in /home/mstorti/FASTQ_Rabbit/run1_2/;
do
samp=`basename ${fn}`
echo "Processing sample ${samp}"
salmon quant -i /home/mstorti/FASTQ_Rabbit/run1_2/salmon/index
         -l A \
         -1 ${fn}/${samp}_R1_001.fastq \
         -2 ${fn}/${samp}_R2_001.fastq \
         -p 8 --validateMappings -o /home/mstorti/FASTQ_Rabbit/run1_2/output_1/${samp}_quant
done

When I submitted the job, the cluster killed it in 20 seconds.

[Pierre Lindenbaum]

the cluster killed it in 20 seconds.

are you using the nodes of cluster or running on a server ?

if 1) you must submit using SGE or SLURM or etc... if 2) you shouldn't run such tool on a small server. See 1

[ATpoint]

First of all, data seems to be in $HOME where on a cluster it typically does not belong. Is there no scratch drive to store data? Then I see uncompressed fastq. No reason for that, gzip it to save space. Also, I see no scheduler instructions in the script, like SLURM as mentioned already. If you're new to cluster computing talk to the admins or a collegaue who did it before. If you run this on the headnode then the kill is expected (and intended).

[m.storti]

You are right. I'm working on a server not on a cluster. I don't know if it is small or not. I will contact the admin. Thank you