Hello, Thank you for providing more data ... very helpful for determining a solution.
The best path would probably be something like this: 1) Re-align the 6 complete D. simulans genomes to dm3. LiftOver can be very useful, but it does come with the disclaimer that it works best lifting one version of *the same* genome to the next version. In addition, one of the input files is the result of a track (Chain) and all of your genomes are not Chain tracks in dm2. Creating your own is possible, but non-trivial. A direct alignment to dm3 would be best. 2) Assuming the you re-aligned to dm3, merging all the alignments should be fairly straightforward. Using the same tool that you used to merge the other alignments for dm3 would be the best option, to keep the methods consistent for the entire experiment. MultiZ is what we use here for the Conservation track MAF alignment merge. I mentioned the dm3 Conservation track because it already contains several fly genomes (plus 2 bee and 1 beetle) mapped to dm3, has highly conserved regions identified (via phastCons), and would be ready to use with no extra prep. After your explanation of the dataset you are using, I agree that it is not an appropriate substitute, but it may be a good comparison track to sanity check any conclusions. Jennifer ------------------------------------------------ Jennifer Jackson UCSC Genome Bioinformatics Group ----- "Yegor Bazykin" <yegor.bazy...@gmail.com> wrote: > From: "Yegor Bazykin" <yegor.bazy...@gmail.com> > To: "Jennifer Jackson" <j...@soe.ucsc.edu> > Cc: "Вова Сеплярский" <pam...@mail.ru>, gen...@soe.ucsc.edu > Sent: Friday, October 16, 2009 3:34:44 AM GMT -08:00 US/Canada Pacific > Subject: Re: [Genome] liftover whole genome > > Hello Jen, > > I have two FASTA files with multiple alignments. One is a set of 37 > complete 2L chromosomes of D. melanogaster genomes (different > strains), > aligned to dm3. The other is a set of 6 complete D. simulans genomes > (different strains), aligned to dm2. I am essentially looking for a > way > to combine the two in a single FASTA file. The current objective is > to > compare the SNPs in each species in homologous (aligned) positions. > > I am not sure how I could use the Conservation track for this > purpose. > > Thanks, > Georgii > > > > Jennifer Jackson wrote: > > Hello Georgii, > > > > Would you be able to add some details: 1) which exact tracks do you > > want to merge (genome and track name). Chain/Net tracks can be > > lifted, true MAF tracks cannot be. If Chain/Net, you could lift and > > then combine using a program like multiZ. 2) have you examined the > > dm3 Conservation track? Would this meet your needs? > > > > If you can help us to understand the experiments/desired results > and > > intended use better, we can help you to determine the best path to > > obtain the data (if possible). > > > > Thanks, Jen > > > > > > ------------------------------------------------ Jennifer Jackson > > UCSC Genome Bioinformatics Group > > > > ----- "Yegor Bazykin" <yegor.bazy...@gmail.com> wrote: > > > >> From: "Yegor Bazykin" <yegor.bazy...@gmail.com> To: > >> gen...@soe.ucsc.edu Cc: "Вова Сеплярский" <pam...@mail.ru> Sent: > >> Thursday, October 15, 2009 4:57:40 AM GMT -08:00 US/Canada Pacific > > >> Subject: Re: [Genome] liftover whole genome > >> > >> Hi all, > >> > >> I have two whole-genome multiple alignments of Drosophila species > >> that I would like to combine. The problem is that one alignment is > >> based on dm2, while the other is based on dm3. I understand that > >> the mapping of the coordinates from dm3 to dm2 is given in > >> dm3ToDm2.over.chain, and that I can use liftOver to convert sets > of > >> coordinates from one assembly to the other. But I don't see an > easy > >> way to combine whole alignments based on two different assemblies. > >> (I ultimately need the alignments in fasta format.) > >> > >> I could write a utility that would use chain files to combine the > >> alignments. But is there a better way to proceed? > >> > >> Thanks, Georgii Bazykin > >> > >> _______________________________________________ Genome maillist - > >> Genome@lists.soe.ucsc.edu > >> https://lists.soe.ucsc.edu/mailman/listinfo/genome > > _______________________________________________ Genome maillist - Genome@lists.soe.ucsc.edu https://lists.soe.ucsc.edu/mailman/listinfo/genome
