freeman-lab / starfish

*fish -- a standardized analysis pipeline for image-based transcriptomics

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

The goal of starfish is to prototype a reference pipeline for the analysis of image-based transcriptomics data that works across technologies. This is a work in progress and will be developed in the open.

concept

See this document for details. The diagram below describes the core pipeline components and associated file manifests that this package plans to standardize and implement.

alt text

usage

Install the starfish package:

% pip install starfish

See the notebook for a fully worked example.

You can also re-produce the notebook results with the command line tool. For usage, run starfish --help

Running the commands below will re-produce notebook results.

mkdir -p /tmp/starfish/raw
mkdir -p /tmp/starfish/formatted
mkdir -p /tmp/starfish/registered
mkdir -p /tmp/starfish/filtered
mkdir -p /tmp/starfish/results

python examples/get_iss_data.py /tmp/starfish/raw /tmp/starfish/formatted --d 1

starfish register /tmp/starfish/formatted/org.json /tmp/starfish/registered --u 1000

starfish filter /tmp/starfish/registered/org.json /tmp/starfish/filtered/ --ds 15

starfish show /tmp/starfish/filtered/org.json

starfish detect_spots /tmp/starfish/filtered/org.json /tmp/starfish/results dots --min_sigma 4 --max_sigma 6  --num_sigma 20 --t 0.01

starfish segment /tmp/starfish/filtered/org.json /tmp/starfish/results stain --dt .16 --st .22 --md 57

starfish decode /tmp/starfish/results --decoder_type iss

visualization quickstart

To see an interactive web-visualization of the final decoded result, run the following commands

  1. Install nvm if you don't have it
  2. Install node nvm install node
  3. Install budo npm install budo
  4. From starfish/viz run npm install
  5. From starfish/viz run npm start

Development

Clone the repo.

% git clone git@github.com:chanzuckerberg/starfish.git
% cd starfish

Virtualenv

While not required, you may wish to set up a virtualenv. If you are using python < 3.3, you should run:

% virtualenv .venv

If you are using python >= 3.3, you should run:

% python -m venv .venv

Installing

Install the starfish module in edit-mode and all the dependencies for starfish:

% pip install -e .
% pip install -r REQUIREMENTS.txt

About

*fish -- a standardized analysis pipeline for image-based transcriptomics

License:MIT License


Languages

Language:Jupyter Notebook 96.0%Language:Python 3.1%Language:JavaScript 0.6%Language:WDL 0.2%Language:Makefile 0.0%