Identifying novel mRNA transcripts

Question

[Pranathi]

After performing, Hisat2-stringTie and DESEQ2 how can I identify if there are any novel mRNA transcripts present in my significant genes or if so which tool do I use? Can someone help me out here? Thanks in advance!

Answer 1

[Kevin Blighe]

Hi, I believe the novel transcripts will be assigned an automatic ID from the HISAT2-StringTie stage, depending on how you have run these programs. Please check the merged GTF that you produce with StringTie to see if any novel transcripts have been identified.

According to a colleague, the automatic ID for novel transcripts may begin with something like 'MSTRG'.

It would help if you showed the commands that you used, and also sample outputs from different stages.

Kevin

[Pranathi]

Oh yes I can check that.

#! /bin/bash
strp=/home/xxx/softwares/stringtie-2.1.4
gtf=/home/xxx/Documents/sss/genome.gtf
thr=40

find ./ -iname "*.bam" >li1
cat li1|while read e
do
        base_name=$(basename $e .bam)
        ${strp}/stringtie -p ${thr} -G ${gtf} -o ${base_name}.gtf -l ${base_name} $e
done

find ./ -iname "*.gtf" >li2.list

${strp}/stringtie -p ${thr} --merge -G ${gtf} -o stringtie_merged.gtf li2.list

cat li1|while read a
do
        base_nam=$(basename $a .bam)
        ${strp}/stringtie -e -p ${thr} -G stringtie_merged.gtf -o ${base_nam}.re.gtf $a
done
find ./ -iname "*.re.gtf"|while read i
do
        aa=$(basename $i .re.gtf)
        printf "$aa\t$i\n"
done >reli2.list #for each path, add SRR accession and space at start of line

python ${strp}/prepDE.py -i reli2.list

This was the total script I have used though.

[Kevin Blighe]

Thanks - I believe they should be found in stringtie_merged.gtf, and then also in your counts data that is produced for DESeq2. It may additionally depend on how you ran HISAT2, though. Nice coding, by the way!

[Pranathi]

Yes, thankyou so much!