Issues with running featureCounts with STAR

Question

[ckim159]

I am trying to generate a count table after aligning my CLIP-seq results to a genome index. I have been using several online tutorials, but I am running into an issue when using featureCounts from bioconda subread. I have tried referencing a folder containing BAM files as well as single BAM files for my input, but I keep getting the same invalid parameters error:

featureCounts -a Mus_musculus.GRCm39.109.gtf -o count.out -T 8 bams/*.bam

ERROR: invalid parameter: 'bams/SRR202.fastqAligned.sortedByCoord.out.bam'

Or when I move the bam files to the folder I am in and try a single input:

featureCounts -a Mus_musculus.GRCm39.109.gtf -o count.out -T 8 SRR202fastqAligned.sortedByCoord.out.bam

ERROR: invalid parameter: 'SRR202.fastqAligned.sortedByCoord.out.bam'

[Ram]

What's the featureCounts version you're using? Can you run featureCounts -h and show the output please?

[ckim159]

featureCounts -h

featureCounts: invalid option -- h
Version 2.0.1

[GenoMax]

That is odd. What is the output of this command file SRR202fastqAligned.sortedByCoord.out.bam?

You should see something like

 gzip compressed data, extra field, original size 0

Edit: Unlike @rfran010 I don't see any problems with the command, switching the order of options etc. Surprisingly *.bam specification picks up all the files automatically (learned something new). I am using v.2.0.3.

[Ram]

I don't think it's a problem with the content of the file or its existence - it's like featureCounts expects a parameter and errors out weirdly.

[rfran010]

It seems like there's an issue with you bam file. How does it look?

on version 2.0.6, if I run your command, I get the same error. If I swap the bam file for one of my own, it will start, give the subread output, but then fail to open the annotation file

[pzm]

Hello,

I am running into a similar issue. I am doing RNA seq on cancer cell lines. Workflow: I got back the fastq.gz files, used trimmomatic for trimming, then mapping with STAR, then conversion to BAM, then sorting, all without any issues reported, and then tried running featurecounts. I have done the same thing for various samples in one go, but with this bam file (and other bam files from this particular cell line) I have the following issue:

commandline:

sbatch -n 1 --mem-per-cpu=8000 -c 10 featureCounts_90-8.sh

.sh file:

#!/bin/bash
echo "Start - `date`"
START=$(date +%s)
module load SUBREAD
featureCounts -a Homo_sapiens.GRCh38.105.gtf -p -o 90-8_D21_R2_featurecounts 90-8_D21d_R2_sorted_Aligned.out.bam
ls -R /etc > /tmp/x
rm -f /tmp/x

Slurm content:

Start - Tue Aug 22 16:35:03 BST 2023
[+] Loading SUBREAD 2.0.3
ERROR: invalid parameter: '90-8_D21d_R2_sorted_Aligned.out.bam'

Again, it has worked for all my files (with other cell lines) except for the files with this cell line.

Does anyone know what might be going on? I am using featureCounts v2.0.3. Any help is much appreciated. Please let me know if you need further information.

BW

[GenoMax]

I wonder if the - in output name is being interpreted as an option. Can you try replacing the - with _ and see if it helps?

I don't remember for v.2.0.3 but for recent version you need to add --countReadPairs in addition to -p to count fragments with PE data.

[Ram]

Are you sure the file 90-8_D21d_R2_sorted_Aligned.out.bam exists? Just going with an educated guess - should the bam file even have R1/R2 in its name?

Answer 1

[pzm]

okay thanks to you I found the issue: I missed a small d for all the files, and so it couldn't find them. Very obvious mistakes. I guess I should have used more easy file names.

I also checked and the - vs _ and it didn't seem to matter.

I should have been able to figure this out by myself, but I checked the file names by eye. Should always do a ctrl F search I suppose. Anyway, Thanks so much for your help!!