I agree that was surprising to me as well. Again, as I said it almost picked the best each time, when it didn't, it was a close 2nd or 3rd, but even the 2nd or 3rd, in my experience is extremely close to the best.
Louis On 13-10-21 10:57 PM, Egon Ozer wrote: > I've never tried this program before, but I'm going to give it a shot. I'm > curious to see how it works, but a little bit skeptical since I've found in > practice that optimal k-mer values vary quite widely from assembler to > assembler. For instance, with my bacterial genomes, I've gotten optimal > assemblies at higher k-mers (51 - 75) with Velvet and lower k-mers (25 - 31) > with Ray using the same set of reads. > > Still, thanks for the referral. > > - Egon > > On Oct 21, 2013, at 9:52 PM, Sébastien Boisvert wrote: > >> On 21/10/13 09:44 PM, Louis Letourneau wrote: >>> We've been doing the many assemblies approach >>> 21,31,41,51,61 >>> Then zooming in say, 43,45,49 >>> >>> This worked well, but uses a lot of cpu time. We are trying out >>> kmergenie and with the last ~6 different assemblies (mostly eukaryotes >>> between 30Mb-2Gb) it's been pretty spot on. >>> >>> http://kmergenie.bx.psu.edu/ >>> Informed and automated k-mer size selection for genome assembly >>> http://bioinformatics.oxfordjournals.org/content/early/2013/08/01/bioinformatics.btt310 >> >> So if I understand correctly, you give your reads to kmergenie and it gives >> you >> back a k value. Then, you use this k value with an assembler that requires >> one. >> Is that so ? >> >>> >>> It takes just a few hours too. >>> >>> I have no interests in the software, I'm just a happy user. >>> >>> Hope this helps >>> Louis >>> >>> On 10/21/2013 5:07 PM, Sébastien Boisvert wrote: >>>> On 21/10/13 04:55 PM, Adrian Pelin wrote: >>>>> Hello, >>>>> >>>>> Thank you for the update. I was wondering, is there any plans in the >>>>> future of optimizing the kmer value? >>>> >>>> No, I don't think there are at the moment. >>>> >>>>> >>>>> I like the idea of SPAdes in which it does multiple assemblies at >>>>> different kmers and then integrates them integrates all assemblies into >>>>> one. >>>>> >>>> >>>> That is a sound thing to do, I should read the SPAdes papers to learn more. >>>> >>>>> Adrian >>>>> >>>>> On 10/21/2013 4:25 PM, Sébastien Boisvert wrote: >>>>>> Dear invaluable community: >>>>>> >>>>>> Ray v2.2.0 was released on 2013-04-17 -- more than 7 months ago. >>>>>> >>>>>> Since this release, there have been steady development (27 closed >>>>>> tickets, >>>>>> 121 commits in Ray, 44 commits in RayPlatform, and so on). >>>>>> >>>>>> The release process was slowed in part due to me writing my thesis >>>>>> and postdoctoral fellowship applications, among other things. >>>>>> >>>>>> >>>>>> Here is a tentative roadmap for upcoming Ray releases: >>>>>> >>>>>> v2.3.0 2013-11-01 >>>>>> v2.3.1 2013-11-25 >>>>>> v2.4.0 2014-01-15 >>>>>> >>>>>> >>>>>> v2.3.0 will be basically a bug fix release with stability improvement and >>>>>> increased user-friendliness. >>>>>> >>>>>> v2.3.1 will a fast release which will focus on lowering memory usage and >>>>>> possibly >>>>>> scaffolding metrics. >>>>>> >>>>>> v2.4.0 will add biom support >>>>>> >>>>>> >>>>>> >>>>>> See https://github.com/sebhtml/ray/issues/milestones >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> Thank you ! >>>>>> >>>>>> P.S. The list has 82 Non-digested Members and 28 Digested Members. >>>>>> >>>>>> ------------------------------------------------------------------------------ >>>>>> October Webinars: Code for Performance >>>>>> Free Intel webinars can help you accelerate application performance. >>>>>> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most >>>>>> from >>>>>> the latest Intel processors and coprocessors. See abstracts and register >>>>>> > >>>>>> http://pubads.g.doubleclick.net/gampad/clk?id=60135991&iu=/4140/ostg.clktrk >>>>>> _______________________________________________ >>>>>> Denovoassembler-users mailing list >>>>>> Denovoassembler-users@lists.sourceforge.net >>>>>> https://lists.sourceforge.net/lists/listinfo/denovoassembler-users >>>>>> >>>>> >>>> >>>> >>>> ------------------------------------------------------------------------------ >>>> October Webinars: Code for Performance >>>> Free Intel webinars can help you accelerate application performance. >>>> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most >>>> from >>>> the latest Intel processors and coprocessors. See abstracts and register > >>>> http://pubads.g.doubleclick.net/gampad/clk?id=60135991&iu=/4140/ostg.clktrk >>>> _______________________________________________ >>>> Denovoassembler-users mailing list >>>> Denovoassembler-users@lists.sourceforge.net >>>> https://lists.sourceforge.net/lists/listinfo/denovoassembler-users >>>> >>> >>> ------------------------------------------------------------------------------ >>> October Webinars: Code for Performance >>> Free Intel webinars can help you accelerate application performance. >>> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most >>> from >>> the latest Intel processors and coprocessors. See abstracts and register > >>> http://pubads.g.doubleclick.net/gampad/clk?id=60135991&iu=/4140/ostg.clktrk >>> _______________________________________________ >>> Denovoassembler-users mailing list >>> Denovoassembler-users@lists.sourceforge.net >>> https://lists.sourceforge.net/lists/listinfo/denovoassembler-users >>> >> >> >> ------------------------------------------------------------------------------ >> October Webinars: Code for Performance >> Free Intel webinars can help you accelerate application performance. >> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most >> from >> the latest Intel processors and coprocessors. See abstracts and register > >> http://pubads.g.doubleclick.net/gampad/clk?id=60135991&iu=/4140/ostg.clktrk >> _______________________________________________ >> Denovoassembler-users mailing list >> Denovoassembler-users@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/denovoassembler-users > > > ------------------------------------------------------------------------------ > October Webinars: Code for Performance > Free Intel webinars can help you accelerate application performance. > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from > the latest Intel processors and coprocessors. See abstracts and register > > http://pubads.g.doubleclick.net/gampad/clk?id=60135991&iu=/4140/ostg.clktrk > _______________________________________________ > Denovoassembler-users mailing list > Denovoassembler-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/denovoassembler-users > ------------------------------------------------------------------------------ October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60135991&iu=/4140/ostg.clktrk _______________________________________________ Denovoassembler-users mailing list Denovoassembler-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/denovoassembler-users
