Please use the mailing list at https://lists.sourceforge.net/lists/listinfo/denovoassembler-users
> From: matteo brilli [matteo.bri...@fmach.it] > Sent: Monday, May 19, 2014 6:58 AM > To: Boisvert, Sebastien > Subject: technical information > > > > > Dear Sebastien, > > I started using Ray a few months together with Abyss for assemblying a paired > end illumina library of a genome we know little (e.g. if it is highly > repetitive or not, even the number of genetically different individuals is > not sure being impossible to cultivate > this organism separately from its host). So, let's say we did the assemblies. > Now, with abyss (more or less all default params) we get larger assemblies > than with Ray and better N50; the main difference is the "redundancy" of > these assemblies. With the term > redundancy I mean the fact that in Abyss much more than in Ray, we get some > of the contigs that are joined together multiple times to give the final > assembly. Can you define what you mean by redundancy ? Do you mean that you are seeing duplicated sequences in your assembly (in this case, with ABySS) ? > It is true that abyss was run for k=60 to 95 in step of 5, I don't know what is "step of 5". > so we have little overlapping of the same > kmers for the two softwares, anyway where there is overlapping what I am > saying above is true and it is evident considering that for the abyss > assemblies, when I group all gene predictions based on homology at 99% > identity I end with N groups = 0.6 N genes, > while in Ray doing the same the groups are 98% of the number of genes. > Hypothesis: you said you have more contigs with Ray. So maybe some genes from your Ray assembly are in more than 1 contigs and that generates more groups to show up. > I don't know if the redundant part in abyss is the result of assemblying > errors whereas Ray is right, but I was wondering how Ray flatten this signal > which is present in the reads. If you want to compare k-mer content between the 2 assemblies, you can actually do that with Ray: mpiexec -n 99 \ Ray -k 31 \ -read-sample-assembly Ray-assembly.fasta \ -read-sample-assembly ABySS-assembly.fasta \ -run-surveyor \ -output Comparison-of-Ray-and-ABySS Just be sure that Ray-assembly.fasta and ABySS-assembly.fasta are multi-line fasta files. > > Cheers > > Matteo > > > > ========================================= > Fondazione Edmund Mach/CRI > Department of Genomics and Biology of Fruit Crops > Functional Genomics Team > Via Mach 1 > 38010 San Michele all'Adige (TN) > Italy > Phone Office: ITA 0461615644 > ========================================= > > > ------------------------------------------------------------------------------ "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE Instantly run your Selenium tests across 300+ browser/OS combos. Get unparalleled scalability from the best Selenium testing platform available Simple to use. Nothing to install. Get started now for free." http://p.sf.net/sfu/SauceLabs _______________________________________________ Denovoassembler-users mailing list Denovoassembler-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/denovoassembler-users
