Effectiveness of model-based clustering in analyzing Plasmodium falciparum RNA-seq time-course data [version 1; peer review: 1 approved, 1 approved with reservations, 1 not approved]

Background: The genomics and microarray technology played tremendous roles in the amount of biologically useful information on gene expression of thousands of genes to be simultaneously observed. This required various computational methods of analyzing these amounts of data in order to discover information about gene function and regulatory mechanisms. Methods: In this research, we investigated the usefulness of hidden markov models (HMM) as a method of clustering Plasmodium falciparum genes that show similar expression patterns. The Baum-Welch algorithm was used to train the dataset to determine the maximum likelihood estimate of the Model parameters. Cluster validation was conducted by performing a likelihood ratio test. Results: The fitted HMM was able to identify 3 clusters from the dataset and sixteen functional enrichment in the cluster set were found. This method efficiently clustered the genes based on their expression pattern while identifying erythrocyte membrane protein 1 as a prominent and diverse protein in P. falciparum. Conclusion: The ability of HMM to identify 3 clusters with sixteen functional enrichment from the 2000 genes makes this a useful method in functional cluster analysis for P. falciparum