# scry
[![Build Status](https://travis-ci.com/kstreet13/scry.svg?token=o1x5ZKVR5sA6MpqhDnQX&branch=master)](https://travis-ci.com/kstreet13/scry)
[![codecov](https://codecov.io/gh/kstreet13/scry/branch/master/graph/badge.svg?token=2QCzltvkbJ)](https://codecov.io/gh/kstreet13/scry)
<img src=inst/scry_sticker.png height="200">
The released version of scry can be found on Bioconductor at https://bioconductor.org/packages/scry.
A collection of methods for the analysis of small-count data, such as single-cell RNA-seq.
Included methods:
- Deviance for feature selection
- GLM-PCA for dimension reduction
- Null residuals, a transformation that when combined with PCA approximates GLM-PCA
Planned additions for upcoming release:
- Quasi-UMI normalization for single-cell read counts
- Cell type prediction for scRNA-seq
## Installation
```
if (!requireNamespace("BiocManager", quietly=TRUE))
install.packages("BiocManager")
BiocManager::install("scry")
```
The Bioconductor release requires R version 4.0. If you prefer to use the older
R 3.6, you can install from the unofficial r3 branch as follows:
```
remotes::install_github("kstreet13/scry@r3")
```
## Issues and bug reports
Please use https://github.com/kstreet13/scry/issues to submit issues, bug reports, and comments.