A comparison and evaluation of five biclustering algorithms by quantifying goodness of biclusters for gene expression data

Abstract

BackgroundSeveral biclustering algorithms have been proposed to identify biclusters, in which genes share similar expression patterns across a number of conditions. However, different algorithms would yield different biclusters and further lead to distinct conclusions. Therefore, some testing and comparisons between these algorithms are strongly required.MethodsIn this study, five biclustering algorithms (i.e. BIMAX, FABIA, ISA, QUBIC and SAMBA) were compared with each other in the cases where they were used to handle two expression datasets (GDS1620 and pathway) with different dimensions in Arabidopsis thaliana (A. thaliana)GO (gene ontology) annotation and PPI (protein-protein interaction) network were used to verify the corresponding biological significance of biclusters from the five algorithms. To compare the algorithms’ performance and evaluate quality of identified biclusters, two scoring methods, namely weighted enrichment (WE) scoring and PPI scoring, were proposed in our study. For each dataset, after combining the scores of all biclusters into one unified ranking, we could evaluate the performance and behavior of the five biclustering algorithms in a better way.ResultsBoth WE and PPI scoring methods has been proved effective to validate biological significance of the biclusters, and a significantly positive correlation between the two sets of scores has been tested to demonstrate the consistence of these two methods.A comparative study of the above five algorithms has revealed that: (1) ISA is the most effective one among the five algorithms on the dataset of GDS1620 and BIMAX outperforms the other algorithms on the dataset of pathway. (2) Both ISA and BIMAX are data-dependent. The former one does not work well on the datasets with few genes, while the latter one holds well for the datasets with more conditions. (3) FABIA and QUBIC perform poorly in this study and they may be suitable to large datasets with more genes and more conditions. (4) SAMBA is also data-independent as it performs well on two given datasets. The comparison results provide useful information for researchers to choose a suitable algorithm for each given dataset.

REW-ISA: unveiling local functional blocks in epi-transcriptome profiling data via an RNA expression-weighted iterative signature algorithm

“… where l is the number of GO terms that enriched in each LFB, m non is the number of genes covered by LFB but not enriched by any GO term. The higher the WE_score is, the more biologically significant the LFBs are [ 55 ]. Three state of the art algorithms, Xmotifs [ 36 ], Bimax [ 37 ], ISA [ 35 ] were conducted for biclustering on real data in comparison to REW-ISA.…”

Analysis of miRNA expression profiles in breast cancer using biclustering

“…According to recent surveys [ 38 , 39 ] that compare several biclustering algorithms on gene expression datasets, the ISA algorithm [ 34 ] is one of the most effective. Given a gene expression matrix, ISA algorithm discovers simultaneously sets of coregulated genes and the corresponding sets of experimental conditions.…”

Scatter search-based identification of local patterns with positive and negative correlations in gene expression data

“…These algorithms can be classified according to the type of patterns that are found, the size of the biclusters or the heuristic strategies that are used [8,15,16]. There is not a common criterion to compare different algorithms [17,18]. Some comparison methodologies are based on statistical metrics [19] or on the study of the behaviour over known synthetic data sets [20].…”