HomePage

This is the supplementary data for Martin, Gloor, Dunn, Wahl (Bioinformatics)

You can move to the supplementary data for Gloor, Martin, Wahl and Dunn (Biochemistry. 2005 May 17;44(19):7156-65) by clicking this link.

This Table1 of the residues that share MI with only one other from the alignments that contain at least 125 sequences.

The MIvsH page contains further information on the relationship between MI and entropy, as well as how different evolutionary models affect that relationship.

The ContactvsRank page contains information about percentile scores of normalized MI

This site contains information on the alignments that contain > 125 sequences. There are a a number of summary files, a picture of a representative structure associated with each entry, and a connection graph. The figures have the following format:

• conserved residues in shades of blue, darkest blue is for most conserved, lightest blue is for least conserved (corresponds to an entropy of 0.3 (base20) or an information content of 3 bits). MI was not calculated for these residues.
• residues with MI scores > 4 are shown in red if the both of the residues involved shares MI with only each other, in orange if the residue shares MI with a small number of others (< 3), and in yellow if the residue shares MI with 3 or more others.
• connection graphs are shown. The numbering inside the circle corresponds to the residue number in the pdb file. Numbering to the left of the connecting line corresponds the MI Z score (not the distance!)

• FirstList - /*These protein families contain >= 125 sequences, and have at least one pair of residues where both residues coevolve only with one other and the coevolving residues are separated by 10 or more others. These families were used in the calcul

ation for number of sequences vs distance given in Table 1. They were also used the calculations reported in Table 2.*/

• SecondList - */These contain >= 125 sequences in which coevolving single residues are separated by less than 10 residues, or where no single coevolving residue pairs are found. These families were not used in the calculation for number of sequences vs d

istance for Table 1, but these were used in the calculations reported in Table 2./*

• MasterList of MI calculations on all alignments from the January 2005 cdd database, including those with < 125 sequences.

• Alignments and residue frequencies from the ATP synthase subunits epsilon and gamma. These were used for the calculations reported in panel B of Figure 2.

*note: these have been tested on OS X only. They should run on any *ix-type machine, type the name of the file to get help after installing the programs

• MI.pl calculates raw MI from a fasta formatted MSA
• peerZ_4.pl calculates the Z score from

the MI file

• graph.pl draws a graph of significant Z

score connections suitable for viewing in Graphviz