Email updates

Keep up to date with the latest news and content from BMC Bioinformatics and BioMed Central.

This article is part of the supplement: 22nd International Conference on Genome Informatics: Bioinformatics

Open Access Open Badges Proceedings

MUCHA: multiple chemical alignment algorithm to identify building block substructures of orphan secondary metabolites

Masaaki Kotera1, Toshiaki Tokimatsu1, Minoru Kanehisa12 and Susumu Goto1*

Author affiliations

1 Bioinformatics Center, Institute for Chemical Research, Kyoto University, Uji, Kyoto 611-0011, Japan

2 Human Genome Center, Institute of Medical Science, University of Tokyo, 4-6-1 Shirokanedai, Minato-ku, Tokyo 108-8639, Japan

For all author emails, please log on.

Citation and License

BMC Bioinformatics 2011, 12(Suppl 14):S1  doi:10.1186/1471-2105-12-S14-S1

Published: 14 December 2011



In contrast to the increasing number of the successful genome projects, there still remain many orphan metabolites for which their synthesis processes are unknown. Metabolites, including these orphan metabolites, can be classified into groups that share the same core substructures, originated from the same biosynthetic pathways. It is known that many metabolites are synthesized by adding up building blocks to existing metabolites. Therefore, it is proposed that, for any given group of metabolites, finding the core substructure and the branched substructures can help predict their biosynthetic pathway. There already have been many reports on the multiple graph alignment techniques to find the conserved chemical substructures in relatively small molecules. However, they are optimized for ligand binding and are not suitable for metabolomic studies.


We developed an efficient multiple graph alignment method named as MUCHA (Multiple Chemical Alignment), specialized for finding metabolic building blocks. This method showed the strength in finding metabolic building blocks with preserving the relative positions among the substructures, which is not achieved by simply applying the frequent graph mining techniques. Compared with the combined pairwise alignments, this proposed MUCHA method generally reduced computational costs with improving the quality of the alignment.


MUCHA successfully find building blocks of secondary metabolites, and has a potential to complement to other existing methods to reconstruct metabolic networks using reaction patterns.