MADIBA: a web server toolkit for biological interpretation of Plasmodium and plant gene clusters

Abstract

Background: Microarray technology makes it possible to identify changes in gene expression of an organism, under various conditions. Data mining is thus essential for deducing significant biological information such as the identification of new biological mechanisms or putative drug targets. While many algorithms and software have been developed for analysing gene expression, the extraction of relevant information from experimental data is still a substantial challenge, requiring significant time and skill.

Description: MADIBA (MicroArray Data Interface for Biological Annotation) facilitates the assignment of biological meaning to gene expression clusters by automating the post-processing stage. A relational database has been designed to store the data from gene to pathway for Plasmodium, rice and Arabidopsis. Tools within the web interface allow rapid analyses for the identification of the Gene Ontology terms relevant to each cluster; visualising the metabolic pathways where the genes are implicated, their genomic localisations, putative common transcriptional regulatory elements in the upstream sequences, and an analysis specific to the organism being studied.

Conclusion: MADIBA is an integrated, online tool that will assist researchers in interpreting their results and understand the meaning of the co-expression of a cluster of genes. Functionality of MADIBA was validated by analysing a number of gene clusters from several published experiments - expression profiling of the Plasmodium life cycle, and salt stress treatments of Arabidopsis and rice. In most of the cases, the same conclusions found by the authors were quickly and easily obtained after analysing the gene clusters with MADIBA.

Figure 1

A schematic representation of the flow of data through MADIBA. After a microarray experiment, data are normalised and then clustered, since it is hypothesised that the genes in a cluster have common biological implications. A cluster of genes is submitted to MADIBA, either as nucleotide sequences, or gene identifiers. This list of genes can then be subjected to five analysis modules – Gene Ontology Analysis, Metabolic Pathways Analysis, Transcription Regulation Analysis, Chromosomal Localisation Analysis and an Organism Specific Analysis. Also shown are the data that are required by each of the analysis modules. The results from the analyses can be exported as a PDF file, or as plain text.

Figure 2

Results from the Metabolic Pathways module. Analysis of cluster 6 of the Plasmodium data [40] revealed that it was noticeably involved in glycolysis. In the KEGG map for glycolysis, it could be seen that almost all of the enzymes involved are present in the cluster. Additionally, all of the enzymes were annotated by all the three annotation sources – the curated annotation from the original data source (PlasmoDB); the semi-automatic KEGG annotation and the automatic PRIAM annotation, as indicated by the yellow boxes.

Figure 3

Results from the Gene Ontology module. An analysis of the biological process ontology of the cluster 6 of the Plasmodium data [40] revealed that anaerobic glycolysis was the most significant term. The DAG was reduced to show only the terms that are most relevant to glucose metabolism. The grey ellipses contain the genes that are annotated to the connected GO term and the colour of the GO terms indicates different levels of significance, as indicated by the legend.

Figure 4

Results from the Arabidopsis data. (A) Analysis of cluster 0 from the Arabidopsis salt stress experiment [42] with the Metabolic Pathways module revealed that the cluster contained genes involved in lignin biosynthesis. The red colour indicates that the annotations were found by two annotation methods (PRIAM and KEGG in this case), and the purple indicates the enzyme was annotated by PRIAM only. (B) After analysing cluster 8 of the Arabidopsis data [42] with the Transcription Regulation module, it was possible to identify putative transcription factor binding sites. The output of the oligo-analysis tool of RSAT is shown, indicating two motifs on the reverse complement that were identified as similar to the WRKY binding site ((C/T)TGAC(T/C)) (highlighted in the red box). Cluster 8 is known to contain several WRKY transcription factors and several disease-resistance genes. (C) Output from the Patch program of the TRANSFAC sub-module. Shown is the PR-1a (a pathogenesis related protein) promoter binding site that was identified. The table headers are provided for convenience.