Hi Jacob, Principal component analysis (pca) might work but it will modify the original dimensions as a linear combination, which might make it difficult to derive physical meaning out of it. May be, you could use decision trees which give results that are easier to interpret. Looking at decision trees can also help in the analysis of pca results. Easy to use implementation of decision trees can be found in weka software- http://www.cs.waikato.ac.nz/ml/weka/
Regards, Harsh On Wednesday, July 8, 2015, Tim Gruene <[email protected]> wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Dear Jacob, > > information about Thomas Schneider's program 'escet' that you remember > can be found at > http://www.embl-hamburg.de/~tschneider/escet/index.html > > Best, > Tim > > On 07/08/2015 03:03 AM, Keller, Jacob wrote: > > Is anyone aware of a way to classify large numbers (100s) of > > conformationally-diverse crystal structures of a single protein > > (here calmodulin)? Pairwise RMSD matrixes seem possible, but may be > > complicated since there are two somewhat stable lobes, and the > > flexible linker in the middle. What I am imagining is a sort of > > multidimensional tree depicting the relationships in conformation > > space of the various structures. > > > > I remember something for this called esct or similar, but can't > > seem to google it. > > > > Any thoughts? > > > > Jacob > > > > ******************************************* Jacob Pearson Keller, > > PhD Looger Lab/HHMI Janelia Research Campus 19700 Helix Dr, > > Ashburn, VA 20147 email: [email protected] <javascript:;> > > ******************************************* > > > > - -- > - -- > Dr Tim Gruene > Institut fuer anorganische Chemie > Tammannstr. 4 > D-37077 Goettingen > phone: +49 (0)551 39 22149 > > GPG Key ID = A46BEE1A > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1 > > iD8DBQFVnMgsUxlJ7aRr7hoRAjWwAKDrqxpq1ixeKNB+mc3h6gDxbJ8RlACePX0P > Ai4HP3/ZY7tcdrPWJCVjhMc= > =rhyn > -----END PGP SIGNATURE----- >
