janxkoci

#!/usr/bin/awk -f

## USAGE
## awk -f est2lgo.awk model_est.txt model.lgo

BEGIN {OFS="\t"}

# read EST file and save param values into array
# note: array includes also header, but that won't match, so nobody cares
NR==FNR {a[$1]=$2; next}

janxkoci

name: base
channels:
  - conda-forge
  - bioconda
dependencies:
## TOOLS
  - mamba # faster conda
  - pip
  - mawk # faster awk
  - miller # data processor

janxkoci

name: admix
channels:
  - conda-forge
  - bioconda
dependencies:
# R packages
  - r-base=4
  - r-rcpp
  - r-tidyverse
  - r-igraph

janxkoci

cd() { builtin cd "$@"; ll; } # Always list directory contents upon 'cd' (macOS style)

janxkoci

#!/bin/bash

## Takes path as argument ($1) or uses current path instead (.) if no argument given.
path="${1:-.}"

## Gets size for items in path and reports the top 10 files and folders.
du -sh ${path}/* | sort -rh | head -n ${2:-10}

janxkoci

Conda cheatsheet

Simple cheatsheet for conda - a package manager designed for scientists and HPCs, that works without root/sudo permissions.

Get conda

The conda manager is available for all platforms in two main versions:

• Anaconda - best for Python-heavy workflows, includes 200+ packages for scientific computing.
• Miniconda - best for other workflows - only the conda manager and its dependencies. It also allows installing any or all packages from Anaconda distribution, and thousands more.

Conda basics

janxkoci

name: bigdata
channels:
  - conda-forge
  - bioconda
dependencies:
  - mawk
  - miller
  - radian # jupyter/julia style R console
  - r-base=4
  - r-essentials

janxkoci

#!/bin/bash

## Launch interactive session for testing purposes (Torque PBS & PBS Pro / OpenPBS)

# set CPUs and RAM (defaults = 8, 16gb)
cores=${1:-8}
ram=${2:-16}

## Torque PBS
qsub -I -N ijob -d . -v PATH -l nodes=1:ppn=${cores},mem=${ram}gb,walltime=48:00:00

janxkoci

name: rbio
channels:
  - conda-forge
dependencies:
  - r-base=4
  - r-essentials
  - r-magrittr
  - r-ape
  - r-seqinr
  - r-poppr

janxkoci

#!/bin/bash

## R in Jupyter Console

## Usage:
# ir.sh     # jupyter-console
# ir.sh qt  # jupyter-qtconsole

## Install Jupyter Console & R kernel (optional: install Jupyter Qtconsole too)
# conda install -c conda-forge jupyter_console r-base r-irkernel # qtconsole