Data

Tools

Indexes

Applications

Experiments

Open Source Science

https://github.com/arogozhnikov/rtg_score

Rank-To-Group score evaluates contribution of confounding factors

https://github.com/arogozhnikov/rtg_score

Science Score: 23.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ○
    codemeta.json file
  • ○
    .zenodo.json file
  • ✓
    DOI references
    Found 1 DOI reference(s) in README
  • ✓
    Academic publication links
    Links to: biorxiv.org
  • ○
    Committers with academic emails
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (13.2%) to scientific vocabulary
Last synced: 11 months ago · JSON representation

Repository

Rank-To-Group score evaluates contribution of confounding factors

Basic Info
  • Host: GitHub
  • Owner: arogozhnikov
  • License: mit
  • Language: Jupyter Notebook
  • Default Branch: master
  • Size: 391 KB
Statistics
  • Stars: 21
  • Watchers: 5
  • Forks: 1
  • Open Issues: 0
  • Releases: 0
Created about 6 years ago · Last pushed over 4 years ago
Metadata Files
Readme License

README.md

Run tests and deploy Run tests and deploy

Rank-To-Group (RTG) score evaluates contribution of confounders

Batch, cell line, donor, plate, reprogramming, protocol — these and other confounding factors influence cell cultures in vitro.

RTG score tracks contribution of different factors to variability by estimating how Rank maps To Group. Scoring relies on ranking by similarity, so there are no explicit or implicit assumptions of linearity.

RTG perfectly works with both well-interpretable data (gene expressions, cell types) and embeddings provided by deep learning.

Usage

rtg_score is a Python package. Installation: bash pip install rtg_score

RTG score requires two DataFrames: one with confounds and one with embeddings (or other features, e.g. gene expressions) ```python from rtgscore import computeRTG_score

following code corresponds to computing an element of the figure above

(exclude same organoid_id and batch+donor)

score = computeRTGscore( metadata=confoundersmetadata, includeconfounders=['batch', 'donor'], excludeconfounders=['organoidid'], embeddings=qpcrdeltact, ) ```

Use compute_RTG_contribution_matrix to compute multiple RTG scores in a bulk .
Example + code for plotting are available in example subfolder.

Parameters

  • metadata - DataFrame with confounding variables
    • sample_id, batch, donor, clone, plate, etc.
  • embeddings - numerical description of samples (DataFrame or 2d array).
    • Examples: qPCR delta Cts, deep learning embeddings, cell types fractions
  • metric - how to define similarity?
    • use euclidean for qPCR and various embeddings and hellinger for cell type distributions
  • included and excluded confounders in example:
    • including ['donor', 'batch'] and excluding ['clone', 'plate'] will estimate how similar are samples with same donor and same batch, while omitting pairs which have same clone or grown on the same plate
    • most use-cases are simple, like include batch effect while exclude plate, but framework is very flexible

Example analysis

Preprint demonstrates application of RTG score to multimodal analysis of cerebral organoids, and demonstrates which conclusions can be drawn.

See example subfolder for an actual code.

Owner

  • Name: Alex Rogozhnikov
  • Login: arogozhnikov
  • Kind: user
  • Location: San Francisco
  • Company: Aperture Science

ML + Science, einops, scientific tools

GitHub Events

Total
Last Year

Committers

Last synced: about 1 year ago

All Time
  • Total Commits: 49
  • Total Committers: 2
  • Avg Commits per committer: 24.5
  • Development Distribution Score (DDS): 0.061
Past Year
  • Commits: 0
  • Committers: 0
  • Avg Commits per committer: 0.0
  • Development Distribution Score (DDS): 0.0
Top Committers
Name Email Commits
Alex Rogozhnikov i****m@g****m 46
Pavan Ramkumar p****r@s****o 3
Committer Domains (Top 20 + Academic)
system1.bio: 1

Issues and Pull Requests

Last synced: about 1 year ago

All Time
  • Total issues: 6
  • Total pull requests: 19
  • Average time to close issues: 2 months
  • Average time to close pull requests: 1 day
  • Total issue authors: 1
  • Total pull request authors: 2
  • Average comments per issue: 0.67
  • Average comments per pull request: 0.05
  • Merged pull requests: 19
  • Bot issues: 0
  • Bot pull requests: 0
Past Year
  • Issues: 0
  • Pull requests: 0
  • Average time to close issues: N/A
  • Average time to close pull requests: N/A
  • Issue authors: 0
  • Pull request authors: 0
  • Average comments per issue: 0
  • Average comments per pull request: 0
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
View more stats
Top Authors
Issue Authors
  • arogozhnikov (6)
Pull Request Authors
  • arogozhnikov (18)
  • pavanramkumar (1)
Top Labels
Issue Labels
Pull Request Labels

Dependencies

setup.py pypi
  • numpy *
  • pandas *
  • scikit-learn *
  • scipy *