Hello Jiahao, We don't expect a 1-1 correspondence between the target and the query, so a reversal will not always recreate the original query. You can read more about the process here: http://genomewiki.ucsc.edu/index.php/Chains_Nets and here: http://www.pnas.org/content/100/20/11484.full
I hope this clears the issue up for you. Please contact us again if you have further inquiries. Best, Antonio Coelho UCSC Genome Bioinformatics Group Jiahao Huang wrote: > Hello, > > I am using liftover to convert hg18 to mm8 using 30bp interval. > everything worked out nicely. But when I convert the output file back to > human coordinate (mm8 coordinate to hg18) using the same method, there > are "unmapped" regions. and when I do it again (now output from the > second conversion - hg18 to mm8), there are still unmapped regions. > > Why is this? I was expecting that the mm8 to hg18 will give me back my > original input coordinate, but instead i lose some information. Isn't > there a one to one coordinate conversion? > > Thanks, > Jiahao > _______________________________________________ > Genome maillist - [email protected] > https://lists.soe.ucsc.edu/mailman/listinfo/genome > _______________________________________________ Genome maillist - [email protected] https://lists.soe.ucsc.edu/mailman/listinfo/genome
