https://github.com/ltla/hvgdetection2018
Justification for the variance modelling and HVG detection methods in scran.
https://github.com/ltla/hvgdetection2018
Last synced: 5 months ago
JSON representation
Justification for the variance modelling and HVG detection methods in scran.
- Host: GitHub
- URL: https://github.com/ltla/hvgdetection2018
- Owner: LTLA
- Created: 2018-10-31T13:41:32.000Z (over 7 years ago)
- Default Branch: master
- Last Pushed: 2019-08-05T01:51:52.000Z (almost 7 years ago)
- Last Synced: 2025-04-05T04:25:59.072Z (about 1 year ago)
- Language: TeX
- Size: 26.4 KB
- Stars: 6
- Watchers: 2
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Description of the HVG machinery in _scran_
This directory contains some files describing the HVG detection machinery in _scran_.
In `theory/`:
- `description.tex` focuses on the theoretical basis behind `trendVar` and `decomposeVar`.
- `comparison.tex` considers the differences between the possible HVG detection methods in _scran_.
In `real/`:
- `fitTrendVar_test.Rmd` tests the behaviour of the `fitTrendVar()` on a range of scenarios.
- `fitTrendCV2_test.Rmd` tests the behaviour of the `fitTrendCV2()` on a range of scenarios.
In `simulations/`:
- `power` contains simulation scripts to compare the performance of `decomposeVar` with `technicalCV2` and `improvedCV2` for detecting HVGs.
Each simulation script describes a different type of variability, though this is not particularly important for gene-wise testing.
- `alpha` assesses the type I error rate for the tests in `decomposeVar`, `technicalCV2` and `improvedCV2`.
- `filter` examines the motivation for the default filter threshold in `trendVar`.