brwnj

"""
qvality must be on the $PATH.
run on a file with pvalues in column 4:

    $ python qvality.py --column 4 input.bed > output.bed

output file will have extra columns for PEP (posterior error prob)
and q-value.

if input.bed has a header, it is retained.

brwnj

#!/usr/bin/env python
# encoding: utf-8
"""
Join R1 and R2, or print unpaired reads of R1 or R2

This script assumes there is whitespace separating the readname from the pair number:

                                       > <
@HW-ST997:140:D0UA5ACXX:8:1101:1497:2138 1:N:0:ATCACG
@HW-ST997:140:D0UA5ACXX:8:1101:1497:2138 2:N:0:ATCACG

brwnj

TODO

• would be nice to be able to send kwargs to run_miso.py
• doesn't account for your paired-end samples at the moment

brwnj

Hela_Pit1_T220A_0
#     Read Sequences:  1937152
#            Aligned:   535159
#   Unique Alignment:   396716
#   Gapped Alignment:    36685
#     Quality Filter:   342465
# Homopolymer Filter:      285

Hela_Pit1_T220A_1
#     Read Sequences:  3214272

brwnj

#! /usr/bin/env python
# encoding: utf-8

import json
import urllib2

# weather underground api key
api_key = ''

u = urllib2.urlopen("http://api.wunderground.com/api/%s/geolookup/q/autoip.json" % api_key)

brwnj

#! /usr/bin/env python
# encoding: utf-8
"""counts of mRNAs per sample where the reads have been aligned to mirbase."""
import sys
import pybedtools
import pandas as pd
import os.path as op
from collections import Counter

def main(args):

brwnj

#! /usr/bin/env bash
#BSUB -J align[1-12]
#BSUB -e gsnap.%J.%I.err
#BSUB -o gsnap.%J.%I.out
#BSUB -q normal
#BSUB -R "select[mem>16] rusage[mem=16] span[hosts=1]"
#BSUB -n 4

<<DOC
align rnaseq samples using gsnap against mm9

brwnj

#!/usr/bin/env bash
#BSUB -J example

JOBIDS=""

# example loop
for (( i = 0; i < 10; i++ )); do
    RUNSCRIPT=example.sh
    echo "#! /usr/bin/env bash" > $RUNSCRIPT
    echo "#BSUB -J example_slave" >> $RUNSCRIPT

brwnj

#!/usr/bin/env bash

n=$#
a=("${@}")
b=("${a[@]:0:$n-1}")
f=${a[$n-1]}
zcat $f | head ${b[*]}

brwnj

# Usage:
# Rscript norm_count_table.R file_in.tab file_out.tab

# source("http://bioconductor.org/biocLite.R")
# biocLite("DESeq")
library(DESeq)

args <- commandArgs(TRUE)
filein <- args[1]
fileout <- args[2]