Sandhya Tiwari, 24 Nov 2015

All modes were calculated using the "./bin/comparative_normalmodes" command on the webnma api. The fasta files were generated using a local installation of MUSTANG
and in the case where a sequence alignment was required, MUSCLE was used.

The alignments were viewed, annotated and managed using JALVIEW v 14.0. All
project files are denoted with a ".jar" extention. 

To prepare the pdb, the files were run through the Polish.py script, which replaces alternative amino acid names etc.

To create a single chain from two or more chains in the PDB, make_single_chain.py was used.

All the TIM barrel domains were defined by CATH, and this annotation is noted
in the excel sheet Dataset_description.xslx in the folder /TIMs_dataset.

Normalised fluctuations were run using the "./bin/comparative_fluctuations"
command on the webnma api, and replotted using the
plot_flucts_from_all_homologues__withnewannotation_080914.r R script in the
/Normalised_fluctuations folder. All the annotations are noted within the
script itself and should be checked against the alignment.


The Deformations energies were calculated using the
"./bin/comparative_deformations" command. To visualise the values, the script
deformations_as_bfactors.py was used to insert them in the B-factor field of
the PDB file. The script requires the deformations energy .dat file and the
PDB file with the CA atoms, which is an output of the webnma normal modes
calculation step.


The correlation analysis is in three parts:

Correlations between pairs of amino acids which are over 8 angstroms apart,
with values that are in the 95th percentile of the scores in the matrix.

Correlations between pairs of amino acids, in networks that either originate
or involve the beta strands, that are over 8 angstroms apart. These correlation values are in the 95th
percentile of the scores in the matrix.

Correlations between pairs of amino acids that are over 4 angstroms apart.
These correlation values are in the 97.5th percentile of the scores in the
matrix.