Question about Read Length parameter during reference creation

[Motoufiq]

Hi,

I’m currently working on the RNA-Seq splicing analysis and have a question about parameter used to build reference STAR (https://github.com/alexdobin/STAR/tree/master) using GENCODE genome (https://www.gencodegenes.org/human/) which then would be related to input parameter for rMATS splicing analysis. My paired end fastq files are 151 bp read length. As this is short read RNA-Seq splicing analysis with 151 bp, I was wondering if I should input --sjdboverhang 150 parameter was to build this STAR reference. NOTE: --sjdboverhang in the STAR specifies the length of the genomic sequence around the annotated junction to be used in constructing the splice junctions database (in my case the PE read length is 151 bp), then my --sjdboverhang 150 (ReadLength-1: 151-1 = 150)

--runMode genomeGenerate
--genomeDir /path/to/genomeDir
--genomeFastaFiles /path/to/genome/fasta
--sjdbGTFfile /path/to/annotations.gtf
--sjdbOverhang ReadLength-1

It seems It’s quite common for Illumina sequencers to produce paired-end reads with a read length of 151 bp instead of the expected 150 bp. Is it OK to build STAR index --sjdboverhang 150? or to build STAR index --sjdboverhang 149 instead?

Additionally, if I don't input --sjdboverhang paramter, while building STAR INDEX, can I run RNA-Seq splicing analysis, and splicing analysis makes sense?

Thank you,
Toufiq