My star alignment setting sjdbOverhang 149 was not accepted by the aligner

[严锋]

Hi 

When I was recently generating the genome index with the new release, I used my old setting for the STAR to generate genome index.

code like this : STAR --runThreadN 16 --runMode genomeGenerate --genomeDir /home/fyan0002/starindex/ensembl87/ --genomeFastaFiles /home/fyan0002/ref_seq/mouse/ENSEMBL/release87/Mus_musculus.GRCm38.dna_sm.primary_assembly.fa --sjdbGTFfile /home/fyan0002/ref_seq/mouse/ENSEMBL/release87/Mus_musculus.GRCm38.87.gtf -sjdbOverhang 149

Because I have the fasta files including 150bp reads, so I set sjdbOverhang to be 149.

The code run successfully and no warning occurred.

When I check the log file, I found the default sjdbOverhang parameter unchanged, still 100.

So I want to seek help with this problem and want to know if it will affect the subsequent alignment step.

Thank you

All the best

[Alexander Dobin]

Hi,

you have one dash in -sjdbOverhang , that is why STAR did not get it.

Cheers

Alex

[严锋]

Hi Alex

Thank you for your reply and apologize for my carelessness.

Another question is about how it will affect the downstream alignment.

If it will make some difference, does that mean that every time I get a sequencing date in FASTA files, I should regenerate the genome index according to the length of reads?

You said in your manual that 100 is an optimal set-up for the parameter. I think it will surely do well with reads less than 100bp, but how about longer reads?

Thanks again for you reply.

Cheers

Feng Yan

在 2017年1月19日星期四 UTC+11上午8:01:03，Alexander Dobin写道：

[Alexander Dobin]

Hi  Feng,

using --sjdbOverjang 100 for reads of any length (>50b) is fine and the difference typically  be negligible.

Please see this post:

https://groups.google.com/d/msg/rna-star/h9oh10UlvhI/BfSPGivUHmsJ

Cheers

Alex