An integrated species comparison analysis based on biological pathways
M. Ovacik1, S. Euling3, B. Sen3, P.G. Georgopoulos21, I.P. Androulakis1
1Rutgers University, 2UDMNJ-RWJMS, 3USEPA
Cross-species comparison often provides insights into underlying principles (how about principles or mechanisms) behind complex biological phenomena1. Evolutionary and organizational relationships among species have been investigated through multiple sequence alignment of a single protein or genome. Yet, comparing genome sequences may not represent the highest (highest) level organization of organisms such as metabolic pathways and protein interaction networks. Comparative analysis of metabolic and signaling pathways allows us to examine complete biological processes rather than individual elements. Bioinformatics approaches such as genomics, proteomics and metabonomics have been used to determine biomarkers for cross-species analysis2. Phylogenetic analysis and bioinformatics approaches, taken together, address the question whether an underlying similarity measure between the mode of action (MOA) i.e. pathways, across species can be quantified. For example, can we assess whether a pathway in human is more similar to its equivalent in mice or rats? In this study, we combine pathway topology, enzyme sequence and promoter similarity of the genes, which code for metabolic enzymes. It is clear that a combined similarity metric will reflect a better evolutionary understanding; .(not clear what you are trying to say here. Maybe rephrase.) pathway topology and enzyme sequence similarity in a given pathway will indicate if the pathways function in the same manner. Furthermore, we explore how likely is that two pathways are regulated in a similar way by comparing relevant promoter regions. We present the similarity trees based on the combined features for each pathway among a large collection of pathways that are constructed in KEGG (I would probably mention the pathways used for this analysis). This procedure may guide us to pick the most appropriate model to study a given biological phenomenon.
1Wachtershauser G.(1990)Proc Natl Acad Sci U S A, 87: 200-4
2Fang H. et al. (2005)BMC Bioinformatics, 6 Suppl 2: S6.