Single-cell integrative Analysis via Latent feature Extraction
Project description
# SCALE v2: Single-cell integrative Analysis via latent Feature Extraction
## Installation #### install from PyPI
pip install scale-v2
#### install from GitHub
git clone git://github.com/jsxlei/scale_v2.git cd scale_v2 python setup.py install
SCALE v2 is implemented in [Pytorch](https://pytorch.org/) framework. Running SCALE v2 on CUDA is recommended if available. Installation only requires a few minutes.
## Quick Start
### 1. Command line
SCALE.py –data_list data1 data2 –batch_categories batch1 batch2
data_list: data path of each batch of single-cell dataset batch_categories: name of each batch
#### Output Output will be saved in the output folder including: * checkpoint: saved model to reproduce results cooperated with option –checkpoint or -c * adata.h5ad: preprocessed data and results including, latent, clustering and imputation * umap.png: UMAP visualization of latent representations of cells * log.txt: log file of training process
#### Useful options * output folder for saveing results: [-o] or [–outdir] * filter rare genes, default 3: [–min_cell] * filter low quality cells, default 600: [–min_gene] * select the number of highly variable genes, keep all genes with -1, default 2000: [–n_top_genes]
#### Help Look for more usage of SCALE v2
SCALE.py –help
### 2. API function
from scale import SCALE adata = SCALE(data_list, batch_categories)
Function of parameters are similar to command line options. Output is a Anndata object for further analysis with scanpy.
#### Tutorial
See document
## Previous version [SCALE](https://github.com/jsxlei/SCALE)
SCALE is still available in SCALE v2
### Command line
SCALE.py –d data –version 1
### API
from scale.extensions import SCALE_v1 SCALE_v1(data)
All the usage is the same with previous SCALE version 1.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for scale_v2-2.0.2.dev0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | fa496a83fdcc85ae6823e68798fe60a655dba1bc13e9a88852d408fe1c1d29f2 |
|
MD5 | f13cb81144bcad6d91b3b4b2d8d1b386 |
|
BLAKE2b-256 | 61625caf38a5f708898222418a084d8b5e85e50217741e84414fad6b5a6828a5 |