Re: [maq-help] map NA18507

[Joel M. <ano...@co...>]

That makes sense,  it also means

fq_all2std.pl seqprb2std

is bugged, at least for pipeline 1.3.1+ data as it gave me the 10 4 10 10...
scores.

Joel


David Dooling wrote:
> On Fri, May 15, 2009 at 11:01:55PM -0700, Joel Martin wrote:
>   
>> the quality numbers look typically precalibration, here's one of mine,
>>
>> before ( prb file to phred score via fq_all2std.pl seqprb2std )
>> 10 4 10 10 10 0 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 7 10
>> 5 10 10 6 10 10 10 10 10 10 10 0
>> and after calibration
>> 29 26 31 31 29 24 27 32 31 30 29 31 27 28 31 31 30 32 28 30 32 32 30 29
>> 30 30 30 28 31 31 30 30 30 27 31 0
>>
>> not sure how you'd fix it, illumina is using a table lately for
>> calibration instead of requiring the phiX
>> values, I think, but i've little to do with that end of things.  Could
>> just lower your thresholds :)
>> Erm, consed was converting from prb to qual in version 17 or 18, and I
>> don't remember seeing
>> scores this low so maybe there's a simple answer.
>>     
>
> I think the simpler answer is that those FASTQ files are using the
> standard (Sanger) offset of 33 (I = 40) rather than the Solexa
> convention of 64 (I = 9).  If I recall correctly, the Solexa pipeline
> tops the quality off at 40 (and the recalibration may too).  It would
> make more sense to have a bunch of high quality bases in a row capped
> at 40 than a bunch of mediocre reads with identical quality in a row.
>
>   
>> kc...@wa... wrote:
>>     
>>> Dear Colleagues,
>>>
>>> I am trying to map the NA18507 reads downloaded from
>>> http://www.ncbi.nlm.nih.gov/Traces/sra/sra.cgi?study=SRP000239
>>> to human reference sequence.  However, I am getting strange mapping
>>> results.  The maximal mapping quality is only 30 and the alternative
>>> mapping quality for a properly mapped pair is only 15.
>>>
>>> e.g.,
>>> SRR003971.245152        1       840373  +       219     18      30      15
>>>      15      1       3       0       1       36
>>> SRR003971.245152        1       840556  -       -219    18      30      15
>>>      15      0       0       1       0       36
>>>
>>> I wonder what I have done wrong?
>>>
>>> Before mapping, I noticed that the fastq files seemed to be in Illumina
>>> format:
>>> @SRR002291.1 FC202P4AAXX_R1:1:1:881:635 length=36
>>> GAAAGAAGTTTTCTGTGAAATGGCTTTGTGATTTGT
>>> +SRR002291.1 FC202P4AAXX_R1:1:1:881:635 length=36
>>> IIIIIIIIIEIIIEI:9:3:IE947IG':.%@6E)8
>>> @SRR002291.2 FC202P4AAXX_R1:1:1:686:750 length=36
>>> GATCGTCTATTACCTTTAAATAACCTAGATCTCAAG
>>> +SRR002291.2 FC202P4AAXX_R1:1:1:686:750 length=36
>>> IIIIIIII<IICIIIII4:3I96197/,,;,;0+*,
>>>
>>> So I converted them using "fq_all2std.pl sol2std".  According to the
>>> instruction these base qualities were only PF filtered but were not
>>> calibrated from alignment.  Would base quality calibration affect the
>>> mapping quality calculations?  How can I make it work?
>>>
>>> Many Thanks.
>>>
>>> -Ken
>>>
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> Crystal Reports - New Free Runtime and 30 Day Trial
>>> Check out the new simplified licensing option that enables 
>>> unlimited royalty-free distribution of the report engine 
>>> for externally facing server and web deployment. 
>>> http://p.sf.net/sfu/businessobjects
>>> _______________________________________________
>>> maq-help mailing list
>>> maq...@li...
>>> https://lists.sourceforge.net/lists/listinfo/maq-help
>>>
>>>   
>>>       
>>
>> ------------------------------------------------------------------------------
>> Crystal Reports - New Free Runtime and 30 Day Trial
>> Check out the new simplified licensing option that enables 
>> unlimited royalty-free distribution of the report engine 
>> for externally facing server and web deployment. 
>> http://p.sf.net/sfu/businessobjects
>> _______________________________________________
>> maq-help mailing list
>> maq...@li...
>> https://lists.sourceforge.net/lists/listinfo/maq-help
>>     
>
>