[Bioperl-l] HSPs: getting the location of mismatch

[Peter Wilkinson]

About the method of EST quality assessment, I am wondering if what your are 
doing is the right way to go.

The first quality assessment that should be made on ESTs should be using 
the trace data to mask poor quality regions on the basis of the actual 
sequencing itself, using something like Phred. Has this been done?

What is the proof-reading step against the genomic data for? Is the source 
of the genomic data really suitable for this? If the trace files were 
assessed for quality, this step is not needed before assembly. I don't see 
this type of alignment as a way of verifying the quality of the ESTs.

It seems to me the quality of the EST sequences are at best, what can be 
obtained from the trace file alone, and all the care that went into the 
experimental prior to sequencing. The Genomic data that is available has 
its own reliability issues ...

Could you explain to me the reasoning (advantages) behind this method or 
perhaps what your goals are? I have not seen it done this way before.



Peter W.

At 01:43 PM 28/04/2003 -0400, Charles Hauser wrote:
>All,
>
>As far as I can tell all programs which return an HSP (blast, blat,
>sim4, exonerate) are all capable of informing you that a mismatch,gap or
>insert occurs in the HSP.
>However, none (as far as I can tell) inform you of the location of the
>mismatch, gap or insert.
>
>Specifically, I'm trying to proofread ESTs against genomic sequence as a
>quality step prior to assembling the ESTs into contigs and would like to
>mask any sequencing errors in the ESTs.
>
>Charles
>
>
>
>_______________________________________________
>Bioperl-l mailing list
>Bioperl-l at bioperl.org
>http://bioperl.org/mailman/listinfo/bioperl-l


-------------------------------------
Peter Wilkinson
Bioinformatics Consultant

-------------------------------------