padilha / biclustlib

A Python library of biclustering algorithms, evaluation measures and datasets.

biclustlib

biclustlib is a Python library of biclustering algorithms, evaluation measures and datasets distributed under the GPLv3 license.

This library is under constant update. We expect to review its code and release a first version soon.

Installation

First you need to ensure that all packages have been installed.

• See requirements.txt
• R >= 3.5;
• biclust R package;
• isa2 R package;
• Other specific libraries may be required by third party implementations that are wrapped in this package;

If you miss something you can simply type:

• pip install -r requirements.txt

If you have all dependencies installed:

• python setup.py install

Citing us

If you use biclustlib in a scientific publication, we would appreciate citations of our paper where this library was first mentioned and used.

To cite biclustlib use: Padilha, V. A. & Campello, R. J. G. B. (2017). A systematic comparative evaluation of biclustering techniques. BMC Bioinformatics, 18(1):55.

For TeX/LaTeX:

@article{padilha2017,
  title={A systematic comparative evaluation of biclustering techniques},
  author={Padilha, Victor A and Campello, Ricardo J G B},
  journal={BMC Bioinformatics},
  volume={18},
  number={1},
  pages={55},
  year={2017},
  publisher={BioMed Central}
}

Available algorithms

Python implementations

• Bi-Correlation Clustering Algorithm (BCCA);
• Bit-Pattern Biclustering Algorithm (BiBit);
• Cheng and Church's Algorithm (CCA);
• Large Average Submatrices (LAS);
• Plaid;
• Conserved Gene Expression Motifs (xMOTIFs);

Wrappers for publicly available software

Python packages

• Factor Analysis for Bicluster Acquisition (FABIA) (wrapper for the pyfabia package);
• Spectral Biclustering (wrapper for the scikit-learn implementation);

R packages

• Binary Inclusion-Maximal Biclustering Algorithm (Bimax) (wrapper for the biclust package);
• Cheng and Church's Algorithm (CCA) (wrapper for the biclust package);
• Plaid (wrapper for the biclust package);
• Iterative Signature Algorithm (ISA) (wrapper for the isa2 package);
• Conserved Gene Expression Motifs (xMOTIFs) (wrapper for the biclust package);

Other implementations

• Bayesian BiClustering (BBC) (wrapper for the executable of the authors' original implementation);
• Binary Inclusion-Maximal Biclustering Algorithm (Bimax) (wrapper for the executable of the authors' original implementation);
• Order-Preserving Submatrix (OPSM) (wrapper for the BicAT software using the executable jar file available here);
• QUalitative BIClustering (QUBIC) (wrapper for the executable of the authors' original implementation);
• RInClose (wrapper for the executable of the authors' original implementation);

All the binaries are available with biclustlib and are compiled for the x86_64 architecture.

Available data collections

• Saccharomyces cerevisiae microarray dataset from Tavazoie et al. (1999) which was used in (Cheng and Church, 2000);
• Saccharomyces cerevisiae and Arabidopsis thaliana microarray datasets used in (Prelić et al. 2006);
• Benchmark of 17 Saccharomyces cerevisiae microarray datasets compiled and preprocessed by Jaskowiak et al. (2013);
• Benchmark of 35 cancer microarray datasets compiled and preprocessed by Souto et al. (2008);

Available external evaluation measures

• The Relative Non-Intersecting Area (RNIA) and Clustering Error (CE) measures proposed by Patrikainen and Meila (2006);
• The recovery and relevance scores proposed by Prelić et al. (2006);
• The match score proposed by Liu and Wang (2007).

Example of use

import numpy as np

from biclustlib.algorithms import ChengChurchAlgorithm
from biclustlib.datasets import load_yeast_tavazoie

# load yeast data used in the original Cheng and Church's paper
data = load_yeast_tavazoie().values

# missing value imputation suggested by Cheng and Church
missing = np.where(data < 0.0)
data[missing] = np.random.randint(low=0, high=800, size=len(missing[0]))

# creating an instance of the ChengChurchAlgorithm class and running with the parameters of the original study
cca = ChengChurchAlgorithm(num_biclusters=100, msr_threshold=300.0, multiple_node_deletion_threshold=1.2)
biclustering = cca.run(data)
print(biclustering)

License (GPLv3)

biclustlib: A Python library of biclustering algorithms and evaluation measures.
Copyright (C) 2017  Victor Alexandre Padilha

This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program.  If not, see <http://www.gnu.org/licenses/>.

biclustlib's Issues

Error in example of use

Hi! when I tried to compile your example I found this error:

from biclustlib.algorithms import ChengChurchAlgorithm

File "/home/alexander/Documents/IA376G/EFC2/biclustlib/biclustlib/algorithms/init.py", line 21, in
from .bcca import BiCorrelationClusteringAlgorithm
File "/home/alexander/Documents/IA376G/EFC2/biclustlib/biclustlib/algorithms/bcca.py", line 21, in
from ._base import BaseBiclusteringAlgorithm
File "/home/alexander/Documents/IA376G/EFC2/biclustlib/biclustlib/algorithms/_base.py", line 23
class BaseBiclusteringAlgorithm(object, metaclass=ABCMeta):
^
SyntaxError: invalid syntax

Any solution or it was my mistake?

Gene names of the datasets in cancer_benchmark

Hello! I have been using the datasets in cancer_benchmark, just wondering if the genes have names instead of just numbers? Thanks!