yjzhang

import numpy as np
from sklearn.cluster.bicluster import SpectralCoclustering

spec = SpectralCoclustering(18)
cluster_counts_subset = np.vstack([cluster_counts[:31, :], cluster_counts[32:,:]])
spec.fit(cluster_counts + 0.0001)
row_labels = spec.row_labels_
column_labels = spec.column_labels_

row_order = np.argsort(row_labels)

yjzhang

# makes an svg bingo board
import svgwrite

def text_to_font_size(text):
    if isinstance(text, str):
        total_text = text
    else:
        total_text = ''.join(text)
    if len(total_text) < 10:
        return 18

yjzhang

#!/usr/bin/env python

# merge DGE matrices from multiple runs of split-seq
# joins on genes...

import os
import sys

import numpy as np
import pandas as pd

yjzhang

#!/bin/bash
# run using 'nohup sh run_split_seq.sh &'

nohup split-seq all --fq1 Matthew_77_S1_R1_001.fastq.gz --fq2 Matthew_77_S1_R2_001.fastq.gz --output_dir output --chemistry v2 --genome_dir /data/reference_genomes/mm10/ --nthreads 16

yjzhang

#!/bin/bash

for i in $(seq 1 18); do
    aws s3 cp --recursive s3://sdfas/S${i}_L1_outputDGE_filtered S${i}_L1_outputDGE_filtered
    aws s3 cp --recursive s3://sdfas/S${i}_L2_outputDGE_filtered S${i}_L2_outputDGE_filtered
done

yjzhang

#!/bin/bash

# kills all gunicorn processes with 8889

pids=`ps ax | grep gunicorn | grep "8889" | awk '{split($0,a," "); print a[1]}'`
# TODO
for pid in $pids; do
    killall -9 $pid
    echo "killed gunicorn process $pid"
done

yjzhang

#!/bin/bash

for var in "$@"; do
    filename=${var%.*}
    #echo $filename
    #echo $@
    echo writing to $filename.html
    pandoc $var --self-contained --toc -c /home/yjzhang/Dropbox/Notes/buttondown.css -s --mathml -o $filename.html
done

yjzhang

def flatten(l):
    output = []
    for x in l:
        try:
            output += flatten(x)
        except:
            output.append(x)
    return output

yjzhang

class TrieNode(object):

    def __init__(self, char = '', children = {}, parent = None, wordEnd = False, value = True):
        self.char = char
        self.children = children
        self.parent = parent
        self.wordEnd = wordEnd
        self.value = value

yjzhang

# simple implementation of quantile normalization? 
import numpy as np 

def quantile_norm(data):
    """
    Source: https://en.wikipedia.org/wiki/Quantile_normalization 

    Note: this doesn't deal with ties very well. 

    Args: