how to get total read count, mapped read count and unmapped read count from multiple sam or bam files

Question

[Bioinfonext]

Hi,

I have got multiple sam files, could anyone please suggest a loop to get the total read count in pair, total mapped read count in pair and total unmapped read count in pair in a tab-delimited file with a file name.

sam file name is like this:

Leaf_T3_S_R2_S45_L001.sam
Leaf_T3_F_R9_S12_L001.sam
Root_T3_F_R7_S29_L001.sam

Many thanks

[Ido Tamir]

you are asking two questions: a) how to loop over files b) how to extract mapped counts. a) is not really a topic for this site and you can find answers on stackoverflow or google. b) was answered multiple times.

[GenoMax]

Use samtools idsxtats.

[Bioinfonext]

Hi genomax,

I have converted sam files to bam and I am trying to use below commnad for getting total unmapped reads number by using reformat.sh from bbtools but it is not giving any output (empty files) ? could you please suggest what is the error?

file names are like this;
Leaf_T1_S_R7_S34_L001.bam
Leaf_T1_S_R8_S33_L001.bam
Leaf_T1_S_R9_S32_L001.bam

batch command:

#!/bin/bash
#SBATCH --job-name=unmppaed_read     # Job name
#SBATCH --partition=k2-lowpri
#########################################################
#RUN_reformat.sh
module add bbtools/38.63 
    reformat.sh -in=Leaf_T1_F_R10_S1_L001.bam out=Leaf_T1_F_R10_S1_L001.unmapped.bam unmappedonly >Leaf_T1_F_R10_S1_L001.unmapped.COUNT.txt
     reformat.sh -in=Leaf_T1_F_R2_S5_L001.bam out=Leaf_T1_F_R2_S5_L001.unmapped.bam unmappedonly >Leaf_T1_F_R2_S5_L001.unmapped.COUNT.txt
     reformat.sh -in=Leaf_T1_F_R3_S4_L001.bam out=Leaf_T1_F_R3_S4_L001.unmapped.bam unmappedonly >Leaf_T1_F_R3_S4_L001.unmapped.COUNT.txt

Thanks

[GenoMax]

There are no - in front of BBTools options. Just in=. I am not sure that redirector will work. Will have to check.

[Bioinfonext]

thanks genomax,

this command is working perfectly;

reformat.sh in=Leaf_T1_F_R10_S1_L001.bam out=Leaf_T1_F_R10_S1_L001.unmapped.bam, unmappedonly

but as you said redirection is not working!

Many thanks

[GenoMax]

You could use a #SBATCH -e file.err in your script, which will collect the stats output for all three jobs in file.err, via SLURM log. You will need to parse out the individual results afterwards.