hanfeisun

"""Library to interface with the GEO (GDS- GEO DataSets) repository"""

import os
import re
import sys
import urllib
import traceback
from datetime import datetime
import time
import enchant

hanfeisun

/*
 * A white-list based PAC without regexp, by @janlay
 * It's just simple and fast.
 * Last update: Dec 9, 2013
 * Special thanks to @Paveo
 */
function FindProxyForURL(url, host) {
    var PROXY = "PROXY 127.0.0.1:8800;SOCKS 127.0.0.1:8801";
    var DEFAULT = "DIRECT";

hanfeisun

first_denom = {1:1,2:5,3:10,4:25,5:50}

def cc_rec(amount, kinds_of_coins = 5):
    if amount == 0:
        return 1
    elif amount < 0 or kinds_of_coins == 0:
        return 0
    else:
        return cc_rec(amount, kinds_of_coins - 1) + \
            cc_rec(amount - first_denom[kinds_of_coins], kinds_of_coins)

hanfeisun

def _macs2(workflow, conf):
    # merge all treatments into one
    merge_bams_treat = ShellCommand(
        "{tool} merge {output[merged]} {param[bams]}",
        tool="samtools",
        input=[target + ".bam" for target in conf.treatment_targets],
        output={"merged": conf.prefix + "_treatment.bam"})
    merge_bams_treat.param = {"bams": " ".join(merge_bams_treat.input)}

    if len(conf.treatment_targets) > 1:

hanfeisun

def _macs2(workflow, conf):
    # merge all treatments into one
    merge_bams_treat = ShellCommand(
        "{tool} merge {output[merged]} {param[bams]}",
        tool="samtools",
        input=[target + ".bam" for target in conf.treatment_targets],
        output={"merged": conf.prefix + "_treatment.bam"})
    merge_bams_treat.param = {"bams": " ".join(merge_bams_treat.input)}

    if len(conf.treatment_targets) > 1:

hanfeisun

def _bowtie(workflow, conf):
    for target in conf.sample_targets:
        bowtie = attach_back(workflow,
            ShellCommand(
                "{tool} -p {param[threads]} -S -m {param[max_align]} \
                {param[genome_index]} {input[fastq]} {output[sam]} 2> {output[bowtie_summary]}",
                input={"genome_dir": os.path.dirname(conf.get_path("lib", "genome_index")),
                       "fastq": target + ".fastq"},
                output={"sam": target + ".sam",
                        "bowtie_summary": target + "_bowtie_summary.txt", },

hanfeisun

CLASS1  = ["BMAL1-ZT10_rep1", "BMAL1-ZT10_rep2"]
CLASS1_PATHS = ["/mnt/Storage/data/exam_data/BMAL1/BMAL1-ZT10_rep1.bam", "/mnt/Storage/data/exam_data/BMAL1/BMAL1-ZT10_rep2.bam"]

CLASS2  = ["BMAL1-ZT22_rep1"]
CLASS2_PATHS = ["/mnt/Storage/data/exam_data/BMAL1/BMAL1-ZT22_rep1.bam"]

GTF = "/mnt/Storage/data/RefGene/mm9.gtf"


SAMPLES = CLASS1 + CLASS2

hanfeisun

def update_one_sample(gsmid, parse_fields=['other_ids', 'paper', 'name', 'species', 'description', 'antibody', 'factor',
                                           'cell type', 'cell line', 'tissue', 'strain']):
    """Given a gsmid, tries to create a new sample--auto-filling in the
    meta fields


    If overwrite is True and there is the sample that has the same gsmid, this function will overwrite that sample

    NOTE: will try to save the sample!!

hanfeisun

#!/bin/bash
sudo date 03050001

# Replace the following line with any application you need to run
open /Applications/Merriam-Webster\ 4.0/Merriam-Webster\ Dictionary.app

# sleep 60

# Set back to the current time using apple service (network needed)
sudo ntpdate -u time.apple.com

hanfeisun

def parseFactorByAntibody(geoPost):
    targetFlds = ["CHIP_ANTIBODY", "ANTIBODY", "CHIP", "ANTIBODY_SOURCE", "ANTIBODY_ANTIBODYDESCRIPTION",
                  "ANTIBODY_TARGETDESCRIPTION"]
    #1. try to get the values
    for t in targetFlds:
        tmp = getFromPost(geoPost, t).strip()
        if not tmp:
            continue
        tmp = tmp.upper().replace("ANTI-", " ").replace("ANTI", " ").strip()
        if len(tmp) < 10 and tmp != "":