cbcrg/psldoc2-nf

Name: psldoc2-nf

Owner: Notredame Lab

Description: null

Created: 2013-10-08 10:02:15.0

Updated: 2016-08-10 08:59:23.0

Pushed: 2016-08-10 08:59:21.0

Homepage: null

Size: 23822

Language: C++

GitHub Committers

UserMost Recent Commit# Commits

Other Committers

UserEmailMost Recent Commit# Commits

README

PSLDoc2

PSLDoc2 is the extended version of PSLDoc and it is implemented based on the nextflow framework. https://github.com/paoloditommaso/nextflow

web site: http://tcoffee.crg.cat/psldoc2

Dependencies
Configuration

Extra R libraries can be specified using the R_LIBS environment variables. Add this variable to the nextflow.config file in the project root folder, as shown in the following example:

env {
  R_LIBS = "$PWD/r_libs/"
}
Stand alone version
Prediction
  1. localizaiton
    • prokaryotic : PSORTb v3 data
    • gram-negative
    • gram-positve
    • archaeal
    • eukaryotic
    • plant protein, http://cal.tongji.edu.cn/PlantLoc/
    • human
    • animal
    • fungi
    • yeast
    • Viral protein
  2. nuclear
    • uclear?????(sub-localization)???????????????????sub-mitochondrian???????
  3. protein function
    • Kinase
Build your own model
  1. training phase

    • psldoc2-train.nf -> CA_train+nFoldValidation.R

    • input: model.fasta

    • output: model.tfpssm, plot_model.json, accuracy-nfold.csv, predict.csv (CA dims)

      $ nextflow psldoc2_train.nf --model=../../../data/PSL/PSORTb3.0/Archaeal.fasta --fold_num=5
      
  2. predicting phase

    • psldoc2-pred.nf -> CA_pred.R (query.fasta, model.tfpssm, CA dims)

    • input: query.fasta, model.tfpssm

    • output: plot_query.json, query.pred

      $ nextflow psldoc2_pred.nf --query small.fa --model data/prokaryotic/Archaeal.tfpssm --CA_dim 36
      

This work is supported by the National Institutes of Health's National Center for Advancing Translational Sciences, Grant Number U24TR002306. This work is solely the responsibility of the creators and does not necessarily represent the official views of the National Institutes of Health.