Журнал протеомики и биоинформатики
Открытый доступ

Абстрактный

Validity of Protein Structure Alignment Method Based on Backbone Torsion Angles

Sunghoon Jung, Se-Eun Bae and Hyeon S. Son

Previous researches noticed that a 3D backbone structure can be mathematically represented with a 1D ? and ? dihedral angle array. However, performance of the backbone dihedral angle alignment was not supported with sufficiently large test sets to be quantified; i.e. only 2 pairs or 4 pairs of proteins were analyzed. Here we showed that it is more effective to accurately anticipate homology among 1891 pairs of proteins of 62 different proteases with the string of ? and ? dihedral angle array than famous 3D structural alignment tool TM-align. Gapless global alignment between protein structures was conducted to validate the effectiveness of performing structural alignment with strings of backbone torsion angles. Representation of 3D structure by 1D torsion angle strings allows local alignment, profile construction, hidden Markov models to be implemented with minor modifications and with almost no loss of speed compared with sequence alignment. By our further validation from the previous studies, the utility of backbone dihedral angle method could be more evident.