search

itmat / rum

RNA-Seq Unified Mapper
http://cbil.upenn.edu/RUM
MIT License
26 stars 4 forks source link

sam file slight modification #136

Closed greggrant closed 11 years ago

greggrant commented 12 years ago

For the sam file, we currently have the MAPQ field (field five) set 255 in all cases.. 255 means "not available". However this causes gatk to filter out all reads. So I'd like to change the output to put 25 in this field when it's a unique mapper and 0 when it's non-unique.

mdelaurentis commented 11 years ago

Greg,

Do you mean 255 for unique mappers and 0 for non-unique? I was just curious if "25" is a typo or if you want to pick that as an arbitrary quality score.

greggrant commented 11 years ago

I want to pick that (25, not 255) as an arbitrary score. It means rougly 99% accurate. I would calculate it more systematically if there were any information on the real meaning of this quantity, but it appears to be anybody's guess.

On Wed, 10 Oct 2012, Mike DeLaurentis wrote:

Greg,

Do you mean 255 for unique mappers and 0 for non-unique? I was just curious if "25" is a typo or if you want to pick that as an arbitrary quality score.

Reply to this email directly or view it on GitHub: https://github.com/PGFI/rum/issues/136#issuecomment-9308170

mdelaurentis commented 11 years ago

Ok, just checking.

greggrant commented 11 years ago

Actually, one thing I forgot. If the read does not map at all it should get a MAPQ value of zero. So three cases:

Non-Unique: 0 Non mapper: 0 Unique mapper: 25

Thanks

On Wed, 10 Oct 2012, Mike DeLaurentis wrote:

Ok, just checking.

Reply to this email directly or view it on GitHub: https://github.com/PGFI/rum/issues/136#issuecomment-9308337

mdelaurentis commented 11 years ago

Will be fixed in 2.0.3_01