Skip to content

iLearn, a Python Toolkit and Web Server Integrating the Functionality of Feature Calculation, Extraction, Clustering, Feature Selection, Feature Normalization, Dimension Reduction and Model Construction for Classification, Best Model Selection, Ensemble Learning and Result Visualization for DNA, RNA and Protein Sequences.

Notifications You must be signed in to change notification settings

Superzchen/iLearn

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

44 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

update news!!!

  • iLearnPlus - an updated version of iFeature and iLearn is now available! (2021-02-28)

iLearnPlus is the first machine-learning platform with both graphical- and web-based user interface that enables the construction of automated machine-learning pipelines for computational analysis and predictions using nucleic acid and protein sequences. iLearnPlus integrates 21 machine-learning algorithms (including 12 conventional classification algorithms, two ensemble-learning frameworks and seven deep-learning approaches) and 19 major sequence encoding schemes (in total 147 feature descriptors), outnumbering all the current web servers and stand-alone tools for biological sequence analysis, to the best of our knowledge. In addition, the friendly GUI (Graphical User Interface) of iLearnPlus is available to biologists to conduct their analyses smoothly, significantly increasing the effectiveness and user experience compared to the existing pipelines. iLearnPlus is an open-source platform for academic purposes and is available at https://github.com/Superzchen/iLearnPlus/. The iLearnPlus-Basic module is online accessible at http://ilearnplus.erc.monash.edu/.

iLearnPlus-Basic module interface:

Basic module

iLearn: an integrated platform and meta-learner for feature engineering and machine learning analysis and modeling of DNA, RNA and protein sequence data

iLearn is a comprehensive Python-based toolkit, integrating feature extraction/calculation, feature analysis (clustering, feature selection, normalization and dimension reduction), predictor construction, best descriptor/model selection, ensemble learning and performance evaluation for DNA, RNA and protein sequences. iLearn is capable of calculating and extracting a wide spectrum of 18 major sequence encoding schemes that encompass 53 different types of feature descriptors for protein sequences, and also can be used to extract 6 major encoding schemes which encompass 26 and 18 different types of feature descriptors for DNA and RNA sequences. Developed from iFeature, iLearn also integrates six kinds of frequently-used feature clustering algorithms, five feature selection algorithms, and three dimensionality reduction algorithms. Four output feature formats are supported by iLearn, which can be directly used and processed in other tools. Furthermore, five commonly used machine learning algorithms are provided, including SVM (Support Vector Machine), RF (Random Forest), ANN (Artificial Neutral Network), KNN (K-Nearest Neighbours) and LR (Logistic Regression). In order to facilitate users’ interpretability of outcomes, the clustering and dimensionality reduction results generated by iLearn can be further visualized in form of scatter diagrams, while the cross-validation result can be visualized in the form of ROC and PRC curves. This makes iLearn a unique and powerful tool that greatly facilitates feature generation, analysis, training and benchmarking of machine-learning models and predictions.

Installation

  • Download iLearn by
git clone https://github.com/Superzchen/iLearn

iLearn is an open-source Python-based toolkit, which operates depending on the Python environment (Python Version 3.0 or above) and can be run on multi-OS systems (such as Windows, Mac and Linux operating systems). Before running iLearn, user should make sure all the following packages are installed in their Python environment: sys, os, shutil, scipy, argparse, collections, platform, math, re, numpy (1.13.1), sklearn (0.19.1), matplotlib (2.1.0), and pandas (0.20.1). For convenience, we strongly recommended users to install the Anaconda Python 3.0 version (or above) in your local computer. The software can be freely downloaded from https://www.anaconda.com/download/.

For users who want to use iLearn package :

cd to the iLearn folder. All the functions regarding feature extraction, feature or sample clustering and feature selection analysis, normalization, dimension reduction and predictor construction can be executed through sixteen main programs. The sixteen main programs include:

  • iLearn-protein-basic.py Extracting 37 different types of feature descriptors for protein sequences.
  • iLearn-protein-PseKRAAC.py Extracting the 16 types of pseudo K-tuple reduced amino acid composition (PseKRAAC) feature for protein sequence.
  • iLearn-nucleotide-basic.py Extracting 14 different types of feature descriptors for nucleotide sequences.
  • iLearn-nucleotide-acc.py Extracting 6 different types of autocorrelation descriptors for nucleotide sequences.
  • iLearn-nucleotide-Pse.py Extracting 6 different types of pseudo-k-tuple composition descriptors for nucleotide sequences.
  • iLearn-clustering.py Running the feature or sample clustering algorithms.
  • iLearn-feature-normalization.py Running the feature normalization algorithms.
  • iLearn-feature-selectior.py Running the feature selection algorithms.
  • iLearn-dimension-reduction.py Running the dimension reduction algorithms.
  • iLearn-ML-SVM.py Running the SVM algorithm.
  • iLearn-ML-RF.py Running the RF algorithm.
  • iLearn-ML-MLP.py Running the ANN algorithm.
  • iLearn-ML-LR.py Running the LR algorithm.
  • iLearn-ML-KNN.py Running the KNN algorithm.
  • iLearn-descriptor-estimater.py Estimating the prediction ability for the specified descriptors.
  • iLearn-auto-pipline.py Running the iLearn pipeline.

Furthermore, the iLearn package contains other Python scripts to generate the position-specific scoring matrix (PSSM) profiles, predicted protein secondary structure and predicted protein disorder, which have also been often used to improve the prediction performance of machine learning-based classifiers in conjunction with sequence-derived information included in the scripts directory.

Examples to run the scripts in iLearn.

All files in the example commands can be found in the examples directory.

Input and output of iLearn

The input for iLearn is a set of DNA, RNA or protein sequences in a special FASTA format. The FASTA header consists of three parts: part 1, part 2 and part 3, which are separated by the symbol ‘|’. Part 1 is the sequence name. Part 2 is the sample category information, which can be filled with any integer. For instance, users may use 1 to indicate the positive samples and -1 or 0 to represent the negative samples for a binary classification task, or use 0, 1, 2, … to represent the different class in multiclass classification tasks. Part 3 indicates the role of the sample, where e.g. “training” would indicate that the corresponding sequence would be used as the training set for K-fold validation test, and “testing” that the sequence would be used as the independent set for independent testing.

Example 1, Extracting Descriptors for DNA Sequences

Running the following command to obtain the Kmer descriptor:

python iLearn-nucleotide-basic.py --file examples/DNA_training.txt --method Kmer --format svm

Generally, users can get the parameters by specifying the parameter '--help'

Example 2, Feature Analysis Using iLearn

K-Means clustering (kmeans)

python iLearn-clustering.py --file examples/code_for_cluster.txt --method kmeans --sof sample --nclusters 2

Feature normalization

python iLearn- feature-normalization.py --file examples/ DNA_code_training.txt --method ZScore --format svm

Feature selection

python iLearn-feature-selectior.py --file examples/DNA_code_testing.txt --method CHI2 --format svm

Dimension reduction

python iLearn-dimension-reduction.py --file examples/DNA_code_testing.txt --method pca --format svm

Example 3, Predictor Construction Using iLearn

Support Vector Machine (SVM) algorithm

python iLearn-ML-SVM.py --train examples/DNA_code_training.txt --indep examples/DNA_code_testing.txt --auto --format svm --batch 0.5 --out SVM

Example 4, Descriptor and Machine Learning algorithm Performance Evaluation

For a prediction task, the iLearn package can select out the descriptor with the best performance by using the ‘iLearn-descriptor-estimater.py’.

python iLearn-descriptor-estimater.py --config config.txt

Example 5, iLearn pipeline

All the individual functionalities in iLearn can be implemented as a pipeline by using the ‘iLearn-auto-pipeline.py’ script.

python iLearn-auto-pipline.py --config config.txt

For more examples and advanced usage of iLearn, please refer the iLearnManual.pdf for more help.

Citation:

If you find iLearn useful, please kindly cite the following paper:

Zhen Chen, Pei Zhao, Fuyi Li, André Leier, Tatiana T Marquez-Lago, Yanan Wang, Geoffrey I Webb, A Ian Smith, Roger J Daly*, Kuo-Chen Chou*, Jiangning Song*, iFeature: a Python package and web server for features extraction and selection from protein and peptide sequences. Bioinformatics, 2018, 34(14): 2499–2502. https://doi.org/10.1093/bioinformatics/bty140

Zhen Chen, Pei Zhao, Fuyi Li, Tatiana T Marquez-Lago, André Leier, Jerico Revote, Yan Zhu, David R Powell, Tatsuya Akutsu, Geoffrey I Webb, Kuo-Chen Chou, A Ian Smith, Roger J Daly, Jian Li, Jiangning Song*, iLearn: an integrated platform and meta-learner for feature engineering, machine-learning analysis and modeling of DNA, RNA and protein sequence data. Briefings in Bioinformatics, 2020, 21(3): 1047–1057. https://doi.org/10.1093/bib/bbz041

Zhen Chen, Pei Zhao, Chen Li, Fuyi Li, Dongxu Xiang, Yong-Zi Chen, Tatsuya Akutsu, Roger J Daly, Geoffrey I Webb, Quanzhi Zhao*, Lukasz Kurgan*, Jiangning Song*, iLearnPlus: a comprehensive and automated machine-learning platform for nucleic acid and protein sequence analysis, prediction and visualization. Nucleic Acids Research , 2021;, gkab122, https://doi.org/10.1093/nar/gkab122

About

iLearn, a Python Toolkit and Web Server Integrating the Functionality of Feature Calculation, Extraction, Clustering, Feature Selection, Feature Normalization, Dimension Reduction and Model Construction for Classification, Best Model Selection, Ensemble Learning and Result Visualization for DNA, RNA and Protein Sequences.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages