DehI internal duplication and CMD family relationships

This is a personal research project that draws on my earlier work as a structural biologist. Although my current research focus is methane modelling, I remain interested in how protein structure constrains—and occasionally enables—major changes in enzyme function. DehI provides a particularly intriguing case in which duplication and fusion may have placed a familiar structural unit into a radically different biochemical context.

DehI is a configuration-inverting 2-haloacid dehalogenase with two structurally similar regions separated by a proline-rich linker. I used the experimentally determined 3BJX structure to divide a 69-sequence DehI family alignment into N- and C-terminal halves, then compared their HH-suite profile hidden Markov models.

The two DehI profiles align across nearly their full lengths with 99.54% HHalign probability, supporting an ancient internal duplication despite only 17% direct sequence identity. A second analysis compares both repeats with compact CMD and AhpD proteins. Clustering intact CMD monomers before profile construction reveals several suggestive CMD subfamily matches that are hidden in the pooled family alignment, although they remain below the threshold for a confident family assignment.

Open the complete illustrated analysis, methods, and downloadable results →