IGV reads (grey vs white)

ocs · 08-03-2011, 12:44 AM

Hello folks,

concerning the Integrative Genomics Viewer:
What is the difference between the white and the grey reads? Some reads in the alignment are all filled with grey and some are outlined with grey but filled with white. I couldn't find any answer in the manual (http://www.broadinstitute.org/softwa.../AlignmentData).

Thanks in advance,
best regards,
Oliver

Seq84 · 08-03-2011, 01:07 AM

Hi,
Reads filled with white have mapping quality score = 0
best regards,
Matteo

ocs · 08-03-2011, 01:28 AM

Thanks!
Does this mean that they are aligned just by desperation?

nilshomer · 08-03-2011, 05:42 AM

It usually means there are two or more equally likely locations, and so this location was randomly chosen from that set.

ocs · 08-03-2011, 05:52 AM

Hello nilshomer,

thanks for your answer. Its a bit confusing, I can think of two equally likly locations that both have a great mapping quality but then its set to zero because its ambiguous?

Jim Robinson · 08-04-2011, 04:27 AM

First apologies this was not documented, we'll correct that. I agree its confusing, I think ambiguous mappings should be indicated by a separate score or field, but that's the way it is.

Jim

Orr Shomroni · 06-13-2012, 07:57 AM

Hello everyone,

To continue on this grey vs. white issue, I have an IGV plot containing 2 reads: 1 long at a certain position, and another slightly shorter which is one position afterwards (i.e. in terms of positions, the small read is contained within the range of the big read). In IGV, the big read is marked white (i.e. can be aligned elsewhere), the small is grey, and the big read has no SNPs/Indels in its edges. The question is, how is it possible that a certain read can be aligned to multiple positions, whereas its subsequence is unique?
On the same note, which tool determines whether a read can be mapped to multiple locations (i.e. white bar)? Can IGV calculate that itself?

Jim Robinson · 06-13-2012, 09:22 AM

Hi,

IGV is not making any judgement calls, is is using the mapping quality. It is a convention with at least some aligners (BWA), if not all aligners, that mapping quality == 0 means the read is not uniquely mapped. IGV does not do any calculation. So if you want to take all interpretation out of the picture a "white" read simply means that the mapping quality has been set to zero by the aligner.

Jim

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Show filtered reads in the IGV?	MBender	Bioinformatics	0	01-25-2012 12:04 PM
Black line between two reads in IGV	kasthuri	Bioinformatics	3	01-13-2012 05:32 AM
RTG Investigator variant detection white paper	Stewart Noyce	Vendor Forum	0	06-30-2011 08:27 AM
missing reads in IGV display	cswarth	Bioinformatics	2	06-28-2011 08:30 AM
Greetings from a grey zone	YohnTheViking	Introductions	0	04-19-2010 12:36 AM

08-03-2011, 12:44 AM	#1
ocs Member Location: Berlin, Germany Join Date: May 2011 Posts: 27	IGV reads (grey vs white) Hello folks, concerning the Integrative Genomics Viewer: What is the difference between the white and the grey reads? Some reads in the alignment are all filled with grey and some are outlined with grey but filled with white. I couldn't find any answer in the manual (http://www.broadinstitute.org/softwa.../AlignmentData). Thanks in advance, best regards, Oliver

06-13-2012, 07:57 AM	#7
Orr Shomroni Member Location: Netherlands Join Date: Oct 2011 Posts: 26	Hello everyone, To continue on this grey vs. white issue, I have an IGV plot containing 2 reads: 1 long at a certain position, and another slightly shorter which is one position afterwards (i.e. in terms of positions, the small read is contained within the range of the big read). In IGV, the big read is marked white (i.e. can be aligned elsewhere), the small is grey, and the big read has no SNPs/Indels in its edges. The question is, how is it possible that a certain read can be aligned to multiple positions, whereas its subsequence is unique? On the same note, which tool determines whether a read can be mapped to multiple locations (i.e. white bar)? Can IGV calculate that itself? __________________ "Though it may seem that all's been said and done, originality still lives on" - some unoriginal guy who had nothing better to write as his signature

08-03-2011, 01:07 AM	#2
Seq84 Member Location: Italy Join Date: Feb 2011 Posts: 19	Hi, Reads filled with white have mapping quality score = 0 best regards, Matteo

08-03-2011, 01:28 AM	#3
ocs Member Location: Berlin, Germany Join Date: May 2011 Posts: 27	Thanks! Does this mean that they are aligned just by desperation?

08-03-2011, 05:42 AM	#4
nilshomer Nils Homer Location: Boston, MA, USA Join Date: Nov 2008 Posts: 1,285	It usually means there are two or more equally likely locations, and so this location was randomly chosen from that set.

08-03-2011, 05:52 AM	#5
ocs Member Location: Berlin, Germany Join Date: May 2011 Posts: 27	Hello nilshomer, thanks for your answer. Its a bit confusing, I can think of two equally likly locations that both have a great mapping quality but then its set to zero because its ambiguous?

08-04-2011, 04:27 AM	#6
Jim Robinson Member Location: Boston, MA Join Date: May 2009 Posts: 75	First apologies this was not documented, we'll correct that. I agree its confusing, I think ambiguous mappings should be indicated by a separate score or field, but that's the way it is. Jim

06-13-2012, 09:22 AM	#8
Jim Robinson Member Location: Boston, MA Join Date: May 2009 Posts: 75	Hi, IGV is not making any judgement calls, is is using the mapping quality. It is a convention with at least some aligners (BWA), if not all aligners, that mapping quality == 0 means the read is not uniquely mapped. IGV does not do any calculation. So if you want to take all interpretation out of the picture a "white" read simply means that the mapping quality has been set to zero by the aligner. Jim