We take the .lav file output from blastz and:
lavToAxt -dropSelf
To get rid of the self alignments. There is also the
kent source program: axtDropSelf
I believe the current version of lastz has options
to avoid the self alignments:
http://www.bx.psu.edu/miller_lab/dist/README.lastz-1.01.50/README.lastz-1.01.50.html#ex_self
http://www.bx.psu.edu/miller_lab/
--Hiram
Kun Gao wrote:
> Dear Sir/Madam,
>
> I am a postdoctor fellow working for the Okinawa Institute of Science
> and Technology of Japan. We are now working on the self-alignment of
> human chromosomes. We did the alignment mainly by the blastz program,
> and then use the axtchain and chainnet program to do the chain and net
> alignments.
>
> But when we were trying to align a gene sequence against itself, for
> example, human chromosome X vs human chromosome X, there were always the
> so-called self-hits in our results. Here we mean by the self-hits a
> piece of gene sequence aligned exactly to itself. We think these results
> are trivial and should be removed. And we found there is no such
> self-hits in the self alignment results downloaded from the UCSC website.
>
> So my questions is, did you do something special to remove the self-hits
> in the alignment?
>
> Thank you very much for your help.
>
> Best regards,
>
> Yours sincerely,
>
> Kun Gao
> _______________________________________________
> Genome maillist - [email protected]
> https://lists.soe.ucsc.edu/mailman/listinfo/genome
>
_______________________________________________
Genome maillist - [email protected]
https://lists.soe.ucsc.edu/mailman/listinfo/genome
