wflynny

# Some universal variables
NCELLS=6000
OUTPUT_NAME="nice-name"
FASTQ_DIR="/path/to/fastqs"
REFERENCE_GENOME="/path/to/reference_dir"
[[ -z "${PBS_NUM_PPN}" ]] && NCORES=20 || NCORES=${PBS_NUM_PPN}

# When reads look like:
# sample-name_S?_L00?_R1_001.fastq.gz
# sample-name_S?_L00?_R2_001.fastq.gz

wflynny

# I usually put this in my ~/.bash_aliases
# A portion of our GPFS storage removes files after 21 days of creation.
# `stat` does not show creation time, so we have to resort to parsing the
# output of `mmlsattr`

ftime() {
    # Usage: 
    #   ftime path/to/file
    # 
    # Outputs:

wflynny

# Find myself referring to this thread a lot:
# https://www.biostars.org/p/198143/
# However updating codes with what I see at JAX
@Mxxxx - MiSeq
@Dxxxx - HiSeq 2500
@Kxxxx - HiSeq 4000
@NSxxx - NextSeq 500/550
@Axxxxx - NovaSeq

wflynny

import scanpy.api as sc
import matplotlib.pyplot as plt
import seaborn as sns

def get_cluster_proportions(adata,
                            cluster_key="cluster_final",
                            sample_key="replicate",
                            drop_values=None):
    """
    Input

wflynny

# Visit the Ensembl ftp site.
# ftp://ftp.ensembl.org/pub/release-95/
#
# You want to find data under the following two URLs:
# 1.  ftp://ftp.ensembl.org/pub/release-95/fasta/[YOUR_SPECIES_HERE]/dna/
# 2.  ftp://ftp.ensembl.org/pub/release-95/gtf/[YOUR_SPECIES_HERE]/
#
# The first file of interest is under the fasta URL:
#     [YOUR_SPECIES_HERE].[ASSEMBLY].dna.primary_assembly.fa.gz
# or, if that doesn't exist,

wflynny

#!/usr/bin/env bash
#### PBS preamble

#PBS -N jupyter-server
#PBS -o /path/to/software/logs/jupyter-server.${PBS_JOBID%%.*}.out

#PBS -j oe
#PBS -m n

#PBS -l mem=128GB

wflynny

_grab_ip() {
  jobid=$1
  port=$2
  hostname=$(qstat -f ${jobid} | grep -oP "exec_host = (\K[a-z0-9]+)")
  echo "http://${hostname}:${port}"
}

_submit_job() {
  queue=$1
  port=$2

wflynny

from sklearn.cluster import KMeans
import numpy as np
import pandas as pd
import scanpy as sc

def load_hto_matrix(mtx_dir):
    raw_htos = sc.read_mtx(mtx_dir + "/matrix.mtx.gz").T
    raw_htos.var = pd.read_csv(mtx_dir + "/features.tsv.gz", header=None, index_col=0)
    raw_htos.obs = pd.read_csv(mtx_dir + "/barcodes.tsv.gz", header=None, index_col=0)
    raw_htos = raw_htos[:, ~raw_htos.var_names.isin(["unmapped"])]

wflynny

#!/usr/bin/env bash

TEMP=$(getopt -o hsg --long help,snapshot,gpu -n 'susuage' -- "$@")

if [ $? != 0 ] ; then echo "Terminating..." >&2 ; exit 1 ; fi

# Note the quotes around `$TEMP': they are essential!
eval set -- "$TEMP"

SNAPSHOT=false

wflynny

import re
import argparse

parser = argparse.ArgumentParser()
parser.add_argument("-i", "--infile", required=True)
parser.add_argument("-o", "--outfile", required=True)
args = parser.parse_args()

gene_matcher = re.compile('\tgene\t.*gene_id (".*?");.*Name (".*?");')
parent_matcher = re.compile('gene_id (".*?");.*Parent (".*?");')