--alignIntronMax parameters setting

[DiracZhu1998]

Dear contributors,
Thank you for giving us a wonderful RNA alignment tools, STAR, hooray!
I have a question about default --alignIntronMax parameters (2ˆwinBinNbits)*winAnchorDistNbins),
how do you set the default value of winBinNbits and winAnchorDistNbins?
What dataset you based on to conclude?
If I calculate a intron distribution and choose 99 percentile as --alignIntronMax parameter (max gap open length I summarize) is OK or not ?
The intron distribution file attached, you may want to check.
Thank you for your help!
Intron_distribution_log10.pdf

[alexdobin]

Hi @DiracZhu1998

you can specify directly the value of --alignIntronMax, the other parameters --winBinNbits --winAnchorDistNbins are only used if you do not specify this value.
I would go with the maximum intron length annotated in the species of interest, unless it's outrageously large - which does not seem to be the case judging by your distribution.

Cheers
Alex

[DiracZhu1998]

Great! Thanks!