kdm9

from __future__ import print_function, division
from frisk.kmerhash import *
import screed
from matplotlib import pyplot as plt
from functools import partial
import multiprocessing as mp
from os import path


def seq_ivom(gen):

kdm9

from frisk.kmerhash import *
import screed

r = screed.open('/home/kevin/ws/seqs/TAIR10_gen/TAIR10.fasta.gz')
for x in r:
    seq = x.sequence[:100000]
    break

winsz = 3000
ivs = []

kdm9

# given a sequence find kmers of a given length that occur equal to t or more
# times in a given length of window in the sequence

from collections import deque


def kmer_frequency(text, k, window, thresh=3):
    counts = dict()
    kmers = deque()
    for w in range(len(text)-window+1):

kdm9

#!/usr/bin/env python
from __future__ import print_function
import random
import docopt

DOC = """
USAGE:
    gengen <length> ...

Give a length for each chromosome you want

kdm9

 dpkg-buildpackage -rfakeroot -D -us -uc -i -I
dpkg-buildpackage: source package seqan
dpkg-buildpackage: source version 2.0.0+dfsg-1
dpkg-buildpackage: source distribution UNRELEASED
dpkg-buildpackage: source changed by Andreas Tille <tille@debian.org>
 dpkg-source -i -I --before-build seqan
dpkg-buildpackage: host architecture amd64
 fakeroot debian/rules clean
dh clean --parallel
   dh_testdir -O--parallel

kdm9

fastq=~/ws/gbstrim/data/em_10B8.fq

for ext in 'uncomp' 'gz' 'zstd'
do
	for i in {1..4}
	do
		echo $ext $i
		/usr/bin/time -f 'TIME %U %S %e %M' load-into-counting.py \
			-T 1 \
			-N 4 \

kdm9

Compression;User;Sys;Wall;Rss
uncomp;4.15;0.4;13.67;408252
uncomp;4.19;0.34;22.8;408120
uncomp;4.14;0.37;19.5;408060
uncomp;4.2;0.34;20.34;408284
mean_uncomp;4.17;0.3625;19.0775;408179
gz;9.44;0.1;9.54;408604
gz;9.43;0.12;9.55;408592
gz;9.49;0.12;9.61;408660
gz;9.4;0.14;9.54;408660

kdm9

#include "read_parsers.hh"

int main(int argc, char **argv)
{
    khmer::read_parsers::Read read;
    size_t n_reads = 0;

    if (argc < 2) {
        return 1;
    }

kdm9

#include <seqan/seq_io.h>

using namespace seqan;

int main(int argc, char **argv)
{
    std::string id, seq, qual;

    if (argc < 2) {
        return 1;

kdm9

#include <seqan/seq_io.h>

int main(int argc, char **argv)
{
    std::string id, seq, qual;

    if (argc < 2) {
        return 1;
    }