ShortStack

NanoString ShortStack pipeline for sequencing and variant calling of per molecule short reads.

Note: This program has currently only been optimized to run on the freya serever and the linux platform.

Versions

Please check here for latest stable release.

Getting Started

Make sure you have the following files availble:

• S6 file from imaging in either JSON or CSV format
• FASTA file of reference sequences
• Encoding file that maps basecalls to targets for each pool. A standard file can be found in ShortStack/base_files
• Optional VCF file to create mutation reference sequences if you are searching for a particular set of variants.

Installation

All installation instructions assume you have already installed a working copy of python.

Windows users

If you are trying to run this on Windows (insert sad face) you may first need to install Visual C++ build tools and add cl.ext to your environmental path variables, as discussed here.

1. Install Visual C++ build tools
2. Use the Use the Visual C++ Command Prompt (You can find it in the Start Menu under the Visual Studio folder). Install using pip as detailed below.

Install from github master branch

• stable release, but most likely not the most updated version

To install the lastest development version, from your command line type:

pip install git+https://github.com/summerela/ShortStack.git@master --user

Install from github dev branch

• latest updates, not necessarily tested or stable

To install the lastest development version, from your command line type:

pip install git+https://github.com/summerela/ShortStack.git@dev --user

Required Input:

config.txt

Please see required options here

s6 file:

• must be in JSON format
• CSV files will be converted using Joseph Phan's CSV to JSON converter

encoding.txt

• TSV format
• Must contain following columns:
  • PoolID Target BC
  • PoolID = FOV_x_y coords as Feature ID
  • Target = Target sequence
  • BC = nunberical basecall from imaging

target.fa

• fasta header must be colon delimited
• header must contain columns as follows: id:chrom:start:stop:build:strand
• strand should be entered as either "+" or "-"

Optional Input:

VCF file

• VCF4.3 format
• INFO field must contain STRAND= followed by + or -

To run:

python run_shortstack.py -c path/to/config.txt

Output

• All3spotters: All barcode events
• Bc_counts: Valid HD0s, diversity filtered
• Invalids: Invalid barcodes (color code doesn’t exist in encoding file)
• Rawcounts: HD0/higher depending on cutoff
  • Contains ontargets, offtargets that have a target, cycle/location information
• No_ftm_calls: Features not called
• Ftm_calls: Features called, gene, counts