[Genome] Relation between minMatch and score

[Galt Barber]

Adding -fine seems to help at least some of the alignments.

-Galt


On Fri, 2 May 2008, Galt Barber wrote:

>
> If the two tiles were right next to each other,
> with tileSize 11 and stepSize 5
> they would span 11+5=16.  To get
> a score of 13 I guess it would be
> needing 3 mismatches in the query
> before or after the 16 contiguous
> matching bases that trigger the two tiles,
> perhaps.  Do you want to send us the psl
> output for a specific example?
>
> -Galt
>
>
> On Fri, 2 May 2008, Jessica Wasserscheid wrote:
>
>> Hi,
>>
>> I am running gfClient v33 with tileSize 11 and stepSize 5 with the
>> default minMatch 2.
>> I understand how 2 tiles will trigger an alignment, and the psl output
>> could then report weak alignments with low scores along with a better
>> and bigger alignment for the same sequence.
>> I know the tiles don't have to overlap, but could you explain how I can
>> end up with *only one hit *for a sequence for which the score is 11 or 13 ?
>> I would expect the score to be at least 16, or if it is lower I would
>> expect to have another hit for that sequence.
>>
>> Thanks
>>
>> Jessica Wasserscheid
>>
>> --
>> Jessica Wasserscheid
>> Bioinformatician, Vervet Monkey Genomics Project
>> http://www.genomequebec.mcgill.ca/compgen/submit_db/vervet_project
>> McGill University & Genome Quebec Innovation Center
>> 740 Dr. Penfield Avenue Room 7211
>> Montreal, Quebec Canada, H3A 1A4
>>
>> Email : jessica.wasserscheid at mail.mcgill.ca
>> Phone : (001) 514-398-3311 #00476
>>
>>
>> _______________________________________________
>> Genome maillist  -  Genome at soe.ucsc.edu
>> http://www.soe.ucsc.edu/mailman/listinfo/genome
>>
> _______________________________________________
> Genome maillist  -  Genome at soe.ucsc.edu
> http://www.soe.ucsc.edu/mailman/listinfo/genome
>