MatrixMatchMaker Web interface

MMM finds the sub-matrices of maximum size from two distance matrices. <> The MMM algorithm has been sped up tremendously from the original version and is now available on this server.
This server accepts plain-text FASTA formatted protein sequence files, and will calculate the alignment with MAFFT (if required), distance matrices using the PMB model
For explanation of parameters please read the papers:

The Human Protein Coevolution Network by Tillier, E. R. M and R. L Charlebois (2009) Genome Research

A New, Fast Algorithm for Protein Coevolution Detection using Maximum Compatible Cliques by Alex Rodionov, Alexandr Bezginov, Jonathan Rose, and Elisabeth R.M. Tillier (2011, Algorithms for Molecular Biology 6:17)

MMM requires 2 input files with a minimum of 3 (homologous) sequences each.

Only plain text files in FASTA format!
The FASTA header of each sequence must start with a taxon id, which can be any alphanumerical characters, followed by a | (pipe) symbol.

For example:

>9606|gi:57099 FAKE gene
>YEAST|not a real gene
>Pantoea|gi|291618003|ref|YP_003520745.1| LacI

click here for an actual example

Two MMM results files are output:
A file containing all matchings of the greatest size
A summary table in tab-delimited format

Select matrix files:

      FASTA file 1

      FASTA file 2

Check if files are alignments (in fasta format) and you don't want them realigned

Allowance (between 0 and 1)

Please start with small files (<30), and use an allowance <0.2;
Otherwise your run may take an extremely long time and the page will time-out.
reset form