[Genome] genes structure

[Archana Thakkapallayil]

Hello Thomas,

We do not do any manual curation at UCSC, but rather align the sequences 
as we get them  to the reference genomic sequence using BLAT. You can 
read about the details behind these tracks by pressing on the 
'mini-button' to the left of the actual track display, or by clicking on 
the hyper linked track name in the track controls (below the display).

To generate the EST track, human ESTs from GenBank were aligned against 
the genome using blat. Note that the maximum intron length allowed by 
blat is 750,000 bases, which may eliminate some ESTs with very long 
introns that might otherwise align. When a single EST aligned in 
multiple places, the alignment having the highest base identity was 
identified. Only alignments having a base identity level within 0.5% of 
the best and at least 96% base identity with the genomic sequence are 
displayed in this track.

RefSeq mRNAs were aligned against the human genome using blat; those 
with an alignment of less than 15% were discarded. When a single mRNA 
aligned in multiple places, the alignment having the highest base 
identity was identified. Only alignments having a base identity level 
within 0.1% of the best and at least 96% base identity with the genomic 
sequence were kept.

I hope this information helps you. Please let us know if you have 
further questions.

Regards,

Archana
UCSC Genome Bioinformatics Group

Thomas Lacroix wrote:
> Hi,
> We are determining genes structure using your genome browser. I am 
> wondering what algorithm are you using to align EST or cDNA or Refseq 
> against the genomic sequence ? Is there a manual curation of the 
> results ? What cutoff values are you using ? if you use a local 
> alignment tool (like BLAT, as opposed to a global alignment tool), do 
> you check for exons that are false positive (meaning that alignment has 
> found an "additional" wrong exon somewhere else in the chromosome 
> because it passes the cutoff values) and/or false negative exons 
> (meaning  that the alignment is missing an exon because it didn't pass 
> the cutoff values) ?
> Thank you for your help.
> Best,
> Thomas
>
> _______________________________________________
> Genome maillist  -  Genome at soe.ucsc.edu
> http://www.soe.ucsc.edu/mailman/listinfo/genome
>