Proteases contribute to a broad spectrum of cellular functions. Given a relatively limited amount of experimental data, development of accurate sequence-based predictors of substrate cleavage sites facilitates better understanding of protease functions and substrate specificity. While many protease-specific predictors of substrate cleavage sites were developed, these efforts are outpaced by the growth of the protease substrate cleavage data. In particular, since data for 100+ protease types are available and this number continues to grow, it becomes impractical to publish predictors for new protease types, and instead it might be better to provide a computational platform that helps users to quickly and efficiently build predictors that address their specific needs. To this end, we conceptualized, developed, tested, and released a versatile bioinformatics platform, ProsperousPlus, that empowers users, even those with no programming or little bioinformatics background, to build fast and accurate predictors of substrate cleavage sites. ProsperousPlus facilitates the use of the rapidly accumulating substrate cleavage data to train, empirically assess and deploy predictive models for user-selected substrate types. Benchmarking tests on test datasets show that our platform produces predictors that on average exceed predictive performance of current state-of-the-art approaches. ProsperousPlus is available as a webserver and a stand-alone software package at http://prosperousplus.unimelb-biotools.cloud.edu.au/.
Download and install Anaconda.
Free download from https://www.anaconda.com/download
See https://docs.anaconda.com/free/anaconda/install/index.html
Create the prosperousplus environment.
conda create -n prosperousplus python=3.7
Activate the prosperousplus environment and install the dependencies.
conda activate prosperousplus
pip install -r requirements.txt
Install and configure your java environment.
See https://docs.oracle.com/en/java/javase/17/install/overview-jdk-installation.html
Install and configure your R enviroment.
To get the information the user needs to enter for help, run: python ProsperousPlus.py --help or python ProsperousPlus.py -h
as follows:
cd ProsperousPlus
conda activate prosperousplus
python ProsperousPlus.py -h
usage: it's usage tip.
optional arguments: “-h”, “--help” Show this help message and exit
“--inputType” fasta or peptide.
“--config” The path to the config file.
“--trainfile” The path to the training set file containing the sequences in fasta(peptide) format, where the length of the sequences is 8, 10, 12, 14, 16, 18 or 20.
“--protease” The protease you want to predict cleavage to, eg: A01.001, Or if you want to build a new model, please create a name. There should no space in the model name.
“--outputpath” The path of output.
“--testfile” The path to the test set file containing the sequences in fasta(peptide) format, where the length of the sequences is 8, 10, 12, 14, 16, 18 or 20. If not, it will be divided from the training set.
“--predictfile” The path to the prediction data file containing the sequences in fasta(peptide) format, where the length of the sequences is 8, 10, 12, 14, 16, 18 or 20.
“--mode” Choose the program module to run. Three modes can be used: prediction, TrainYourModel, UseYourOwnModel. Only select one mode each time.
“--modelfile” The path to the trained model generated from the TrainYourModel module. eg 0_model
“--SHAP” Select Yes or No to control the program to calculate SHAP.
“--PLOT” Select Yes or No to control whether the program computes the visualization of cleavage sites.
“--processNum” The number of processes in the program. Note: Integer values represent the number of processes. "processNum" setting can speed up the running efficiency of the program, but it also takes up more computing resources.
cd ProsperousPlus
conda activate prosperousplus
python ProsperousPlus.py --predictfile predict.fasta --outputpath results --inputType fasta --protease A01.001 --mode prediction --PLOT Yes --processNum 2
cd ProsperousPlus
conda activate prosperousplus
python ProsperousPlus.py --trainfile data/A01.001_trainset_1_1.fasta --outputpath resultfile --inputType fasta --protease A01.001 --mode TrainYourModel --SHAP Yes --processNum 2
cd ProsperousPlus
conda activate prosperousplus
python ProsperousPlus.py --predictfile predict.fasta --outputpath resultfile --inputType fasta --protease A01.001 --mode UseYourOwnModel --modelfile modelfile --processNum 2
When the task is prediction or UseYourOwnModel, the result of the program is the test performance of the model; while when the task is TrainYourModel, the result of the program includes the model files, test results, matrix, ROC, SHAP(if selected), and the visualization of cleavage sites (if selected).
shap.summary_plot(shap_values, X_train,max_display=50,show=False, save=True,path='./shap/%s.png'%(d))
Add to the bottom of the summary_plot function:
if save:
pl.savefig(path)
pl.close()This software is for learning and research purposes only. Do not use it for commercial purposes!