From a90d6b2db127d494a0a892de1c79a3bd8a41a15b Mon Sep 17 00:00:00 2001 From: Liang Ding Date: Fri, 9 Aug 2019 11:11:10 -0500 Subject: [PATCH] add missing part of the usage in readme --- README.md | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index e50ca8a..424b3b0 100644 --- a/README.md +++ b/README.md @@ -1,6 +1,4 @@ # SJARACNe -[![Build Status](https://travis-ci.com/jyyulab/SJARACNe.svg?branch=master)](https://travis-ci.com/jyyulab/SJARACNe) - SJARACNe is a scalable solution of ARACNe that dramatically improves the computational performance, especially on the memory usage to allow even researchers with modest computational power to generate networks from thousands of samples. The algorithm uses adaptive @@ -34,7 +32,9 @@ $ conda install --file requirements.txt ### Using pip -First install [Python 3.6.1](https://www.python.org/downloads/) and then use the following command to install SJARACNe and the dependencies. +First install [Python 3.6.1](https://www.python.org/downloads/) and then use the following command to install SJARACNe +and dependencies. + ```$ pip install SJARACNe``` @@ -64,6 +64,13 @@ Subcommands: computing platforms. We have tested it locally using [cwltool](https://github.com/common-workflow-language/cwltool) and on an IBM LSF cluster using [cwlexec](https://github.com/IBMSpectrumComputing/cwlexec). For the convenience, a python wrapper is developed for you to choose computing platform using ```subcommand```. + +The local mode (sjaracne local) runs in parallel by default using cwltool's --parallel option. To run it in serial, +use --serial option. + +To use LSF mode, editing the LSF-specific configuration file SJARACNe/config/config_cwlexec.json to change the default +queue and adjust memory reservation for each step is necessary. Consider increasing memory reservation for bootstrap +step and consensus step if the dimension of your expression matrix file is large. ### Inputs