https://github.com/czbiohub-sf/molecular-cross-validation

Calibrate and compare methods for denoising single-cell RNA-seq data

https://github.com/czbiohub-sf/molecular-cross-validation

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
  • Academic email domains
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (6.1%) to scientific vocabulary
Last synced: 9 months ago · JSON representation

Repository

Calibrate and compare methods for denoising single-cell RNA-seq data

Basic Info
  • Host: GitHub
  • Owner: czbiohub-sf
  • License: mit
  • Language: Python
  • Default Branch: master
  • Homepage:
  • Size: 24.4 MB
Statistics
  • Stars: 2
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Archived
Created over 7 years ago · Last pushed over 5 years ago
Metadata Files
Readme License

README.md

Molecular Cross Validation

Single-cell RNA sequencing enables researchers to study the gene expression of individual cells. In high-throughput methods, however, the portrait of each individual cell is noisy, representing only a few thousand of the hundreds of thousands of mRNA molecules originally present.

Molecular cross-validation is a statistically principled and data-driven approach for estimating the accuracy of any single-cell denoising method without the need for ground-truth. To perform molecular cross-validation, one splits the molecules captured from each cell into two groups, which is equivalent to taking two independent draws from that cell's original mRNA content. The first group is used for model fitting and the second for validation.

In this repository, we show how to correctly calibrate three denoising methods---principal component analysis, network diffusion, and a deep autoencoder---and how to select the best one for a given dataset.

For details about the method and performance on real and simulated data, see the preprint Molecular Cross-Validation for Single-Cell RNA-seq, available here.

Owner

  • Name: Chan Zuckerberg Biohub San Francisco
  • Login: czbiohub-sf
  • Kind: organization
  • Location: San Francisco

GitHub Events

Total
Last Year

Dependencies

setup.py pypi
  • magic-impute *
  • matplotlib *
  • numpy *
  • pandas *
  • scanpy *
  • scikit-learn *
  • simscity *
  • torch *