HISAT2 alignment Error?

Question

[k.kathirvel93]

Hi EveryOne,

I am working in Paired-wnd RNA-seq data set, When i am trying to align my reads against reference HG19 genome by using HISAT2, am getting this error:

Warning: skipping mate #2 of read 'ST-E00142:733:HTKKMCCXY:1:1207:16163:5229 2:N:0:ACTTGA' because it was < 2 characters long

This is happening only after trimming. Thanks in advance.

Answer 1

[ATpoint]

This is not an error but a warning. Apparently your trimmer cuts of most of the read because of an adapter detection or low-quality of the base pairs. It would be smart to discard reads shorter than a certain cutoff, e.g. 20bp categorically as they have little change to be properly aligned (which most trimmers do by default), simply because of the size. The shorter the size, the more likely they will be multimappers. A 2bp read cannot be aligned anyway, as one might imagine quiet well. Please post the full command of the trimming.

[k.kathirvel93]

Thanks for the reply ATpoint, Here is my Adapter removal command:

cutadapt3 -a AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC  -A AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGTAGATCTCGGTGGTCGCCGTATCATT -o Sample_1.fastq -p Sample_2.fastq Sample1_1fastq Sample1_2.fastq

I am not getting any warning, while aligning with raw unfilterd data, with filtered data only am getting warning.

[ATpoint]

Cutadapt has no default minimum read length. I never noticed that because I always use -m 18. Can also be 25 or so. In any case, it should prevent you from these warnings.

[k.kathirvel93]

Should I do something for this warning OR i can go further ?. Because after this, i am getting more than 95 % alignment rate. Thanks

[liya.mangalasseril]

I am also facing the same problem. I have used trimmomatic. Is it ok to proceed with this?