|
From: David E. (gringer) <dav...@mp...> - 2011-07-19 14:35:23
|
So, things are finally looking like the colour-space stuff is going somewhere. I've managed to get a good result from a Ray run with colour-space input (no errors, first bases included), and it might be a tad quicker than the sebhtml/ray colour-space assembler: $ time ../../../sebgit/ray/code/Ray -p tests/phix/phix_5k_1.csfasta tests/phix/phix_5k_2.csfasta | grep -A 7 'Number of contigs:' Number of contigs: 1 Total length of contigs: 5380 Number of contigs >= 500 nt: 1 Total length of contigs >= 500 nt: 5380 Number of scaffolds: 1 Total length of scaffolds: 5380 Number of scaffolds >= 500 nt: 1 Total length of scaffolds >= 500: 5380 real 0m3.107s user 0m2.432s sys 0m0.652s $ time ../code/Ray -p tests/phix/phix_5k_1.csfasta tests/phix/phix_5k_2.csfasta | grep -A 7 'Number of contigs:' Number of contigs: 1 Total length of contigs: 5243 Number of contigs >= 500 nt: 1 Total length of contigs >= 500 nt: 5243 Number of scaffolds: 1 Total length of scaffolds: 5243 Number of scaffolds >= 500 nt: 1 Total length of scaffolds >= 500: 5243 real 0m2.894s user 0m2.392s sys 0m0.484s $ fasta_formatter -i tests/phix/phix.fasta | fastx_reverse_complement | grep $(fasta_formatter -i RayOutput.Scaffolds.fasta | grep -v '^>') > /dev/null && echo "success (match in reverse direction)" success (match in reverse direction) There's nothing particularly impressive with this (apart from it actually working...). For what it's worth, the sebhtml/ray also produces a correct sequence, if you feed it the right starting base: $ fasta_formatter -i tests/phix/phix.fasta | fastx_reverse_complement | grep $((echo -n 'G'; tail -n +2 RayOutput.Scaffolds.fasta) | ~/scripts/cs2base.pl | fasta_formatter) > /dev/null && echo "success (match in forward direction)" success (match in forward direction) However, this is probably more interesting, using a colour-space sequence for one end, and a base-space sequence for the other end: $ time ../../../sebgit/ray/code/Ray -p tests/phix/phix_5k_1.fasta tests/phix/phix_5k_2.csfasta | grep -A 7 'Number of contigs:' Number of contigs: 3 Total length of contigs: 15713 Number of contigs >= 500 nt: 3 Total length of contigs >= 500 nt: 15713 Number of scaffolds: 3 Total length of scaffolds: 15713 Number of scaffolds >= 500 nt: 3 Total length of scaffolds >= 500: 15713 real 0m4.246s user 0m3.428s sys 0m0.796s bioinf@thaliana:~/install/ray/git/ray/system-tests$ time ../code/Ray -p tests/phix/phix_5k_1.fasta tests/phix/phix_5k_2.csfasta | grep -A 7 'Number of contigs:' Number of contigs: 1 Total length of contigs: 5243 Number of contigs >= 500 nt: 1 Total length of contigs >= 500 nt: 5243 Number of scaffolds: 1 Total length of scaffolds: 5243 Number of scaffolds >= 500 nt: 1 Total length of scaffolds >= 500: 5243 real 0m2.892s user 0m2.412s sys 0m0.456s $ fasta_formatter -i tests/phix/phix.fasta | fastx_reverse_complement | grep $(fasta_formatter -i RayOutput.Scaffolds.fasta | grep -v '^>') > /dev/null && echo "success (match in reverse direction)" success (match in reverse direction) [I haven't tried doing a match with the sebhtml/ray sequence.] The commit that actually got my code working properly with colour-space is here: https://github.com/gringer/ray/commit/fbd1efd4c3dbe31df0de820167a068736de648f5 [I forgot to make the same changes for the KmerAcademyBuilder (lines 109-113) that I made for the VerticesExtractor (lines 110-114)] -- David |
