andreas-wilm

#!/usr/bin/env nextflow

/* this is a minimal working example for a bug we recently ran into.
 * in a nutshell: we can function (gen_sample_map_str) in every call to
 * process GenomicsDB.  * the function has static input and should
 * produce identical output each time, but it doesn't. the question is
 * why.
 *
 * yes, we can do tihs different, i.e. construct the string once and yes
 * some things don't make sense in this current minimalistic form.

andreas-wilm

Keybase proof

I hereby claim:

• I am andreas-wilm on github.
• I am andreaswilm (https://keybase.io/andreaswilm) on keybase.
• I have a public key ASCHyaZnBYiR0MNExSduKh5LsVgFwldPfSI1mnHKOYKDZAo

To claim this, I am signing this object:

andreas-wilm

set -euo pipefail

VERSION=2.4.2

pushd /tmp/
wget https://github.com/singularityware/singularity/releases/download/$VERSION/singularity-$VERSION.tar.gz
tar xzf singularity-${VERSION}.tar.gz
cd singularity-${VERSION}
./configure --prefix=/usr/local
make

andreas-wilm

# found at #nfhack17
dig +short myip.opendns.com @resolver1.opendns.com

andreas-wilm

#!/usr/bin/env python

# Example for using repeateable -v and -q for setting logging level
# with argparse. works with python 2 and 3. based on
# https://www.reddit.com/r/Python/comments/3nctlm/what_python_tools_should_i_be_using_on_every/
# 
# masklinn: """A trick I like is using repeatable -v and -q to select
# the logging level. The logging levels are actually integers with a
# 10 increment (DEBUG is 10, CRITICAL is 50).  It's very easy with
# argparse's count action or click's count=True"""

andreas-wilm

from math import log
from scipy import stats

def fisher_comb(pv1, pv2):
    """
    Fisher's method for combining p-values:
    https://en.wikipedia.org/wiki/Fisher's_method

    See for example breseq-0.18b:polymorphism_statistics.r
    

andreas-wilm

from datetime import datetime
def timestamp():
    return datetime.now().isoformat()

andreas-wilm

# assuming:
# - refsq is set and always the same
# - bamfh is an open pysam samfile
for r in bamfh:
    if r.is_unmapped:
      continue

    for (qapos, rpos) in r.get_aligned_pairs():
        # qapos is the aligned index, i.e. this ignores clipping. add that

andreas-wilm

/* -*- c-file-style: "k&r"; indent-tabs-mode: nil; -*- */
#include <stdio.h>
#include <stdlib.h>

#include <time.h>
 
/* returns a timestamp in the form of YYYY-mm-dd HH-MM.
 * buf needs to be able to hold 19+1 characters (incl. \0).
 * returns 0 on success -1 otherwise.
 */

andreas-wilm

import numpy as np
import matplotlib.pyplot as plt

# use the colorblind friendly colors
# https://code.google.com/p/python-cudtools/source/browse/cudtools.py
import cudtools

numlines = 20