""" ProMix: Take a protein structure ensemble and find a mixture of rigid segments or a mixture of conformers. Writes K copies of the ensemble (for segments) or K subsets of the ensemble (for conformers) as PDB files, each superposed on different components. Reference: Hirsch M, Habeck M. - Bioinformatics. 2008 Oct 1;24(19):2184-92 """ import numpy import csb.apps import csb.bio.structure from csb.bio.io.wwpdb import LegacyStructureParser from csb.statistics import mixtures class ExitCodes(csb.apps.ExitCodes): IO_ERROR = 2 class AppRunner(csb.apps.AppRunner): @property def target(self): return ProMixApp def command_line(self): cmd = csb.apps.ArgHandler(self.program, __doc__) cmd.add_scalar_option('components', 'K', int, 'Number of components', -1) cmd.add_scalar_option('type', 't', str, 'Type of mixture', 'segments', ('segments', 'conformers')) cmd.add_positional_argument('infile', str, 'input PDB file') return cmd def initapp(self, args): app = self.target return app(args) class ProMixApp(csb.apps.Application): def main(self): try: parser = LegacyStructureParser(self.args.infile) models = parser.models() except: self.exit('PDB file parsing failed', ExitCodes.IO_ERROR) if len(models) < 2: self.exit('PDB file contains only one model', ExitCodes.USAGE_ERROR) ensemble = parser.parse_models(models) X = numpy.array([model.get_coordinates(['CA'], True) for model in ensemble]) if self.args.type == 'segments': self.main_segments(ensemble, X) elif self.args.type == 'conformers': self.main_conformers(ensemble, X) else: raise ValueError('type must be "segments" or "conformers"') def main_segments(self, ensemble, X): mixture = mixtures.SegmentMixture.new(X, self.args.components) self.log('Number of segments: {0}'.format(mixture.K)) for k,(sigma,w) in enumerate(zip(mixture.sigma, mixture.w)): outfile = 'promix_segment_{0}.pdb'.format(k+1) self.log(' {0}: sigma = {1:6.3f}, w = {2:.3f}, file = {3}'.format(k+1, sigma, w, outfile)) for model, R, t in zip(ensemble, mixture.R, mixture.t): if k > 0: model.transform(R[k-1], t[k-1]) R = R[k].T t = -numpy.dot(R, t[k]) model.transform(R, t) ensemble.to_pdb(outfile) def main_conformers(self, ensemble, X): mixture = mixtures.ConformerMixture.new(X, self.args.components) self.log('Number of conformers: {0}'.format(mixture.K)) membership = mixture.membership for k,(sigma,w) in enumerate(zip(mixture.sigma, mixture.w)): outfile = 'promix_conformer_{0}.pdb'.format(k+1) self.log(' {0}: sigma = {1:6.3f}, w = {2:.3f}, file = {3}'.format(k+1, sigma, w, outfile)) ek = csb.bio.structure.Ensemble() for model, R, t, mk in zip(ensemble, mixture.R, mixture.t, membership): if mk != k: continue R = R[k].T t = -numpy.dot(R, t[k]) model.transform(R, t) ek.models.append(model) ek.to_pdb(outfile) def main(): AppRunner().run() if __name__ == '__main__': main()