Is * really a valid value for a SAM FLAG field?

derobins · 01-19-2011, 03:23 PM

The SAM spec (v1.3-r882, page 3) states that * is a valid value for the SAM FLAG field, yet the table just above it indicates that it is an int in the interval [0, 2^16 - 1] without listing the star (as is done where * is a valid value).

Is the text about * being a valid FLAG value an error? Shouldn't 0 mean that no bits are set, not *?

Thanks,

Dana Robinson
The HDF Group

maubp · 01-20-2011, 10:06 AM

It does sound like a bug in the spec (especially since there is no way to do anything equivalent in BAM). You might want to ask this on the samtools mailing list.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
SAM flag idioms	kgulukota	Bioinformatics	14	08-23-2016 01:32 AM
sam flag is confusing	poorphd	Bioinformatics	6	01-11-2012 11:46 AM
Flag=4 in SAM	Rachelly	Bioinformatics	2	12-22-2010 03:54 AM
SAM flag field and removing unmapped reads from BFAST output	aiden	Bioinformatics	3	05-27-2010 07:10 PM
sam flag 97 and 145	hollandorange	Bioinformatics	8	05-14-2010 03:03 PM

01-19-2011, 03:23 PM	#1
derobins Junior Member Location: Champaign, Illinois Join Date: Jul 2010 Posts: 4	*Is really a valid value for a SAM FLAG field?** The SAM spec (v1.3-r882, page 3) states that * is a valid value for the SAM FLAG field, yet the table just above it indicates that it is an int in the interval [0, 2^16 - 1] without listing the star (as is done where * is a valid value). Is the text about * being a valid FLAG value an error? Shouldn't 0 mean that no bits are set, not *? Thanks, Dana Robinson The HDF Group

01-20-2011, 10:06 AM	#2
maubp Peter (Biopython etc) Location: Dundee, Scotland, UK Join Date: Jul 2009 Posts: 1,543	It does sound like a bug in the spec (especially since there is no way to do anything equivalent in BAM). You might want to ask this on the samtools mailing list.