2
1: RH RK DE AG AP AS AT AV NI NS NT NY CG CF CS CW CY QL QP GS GW GV IL IM IF IS IT IV LM LF LP LS LW LV MT MV FS FY FV PS PT ST SW SY
0: all others

// End of File

Prepared by Wa Yang.  This partitions amino acid changes into the
conserved changes (class 1 above) and radical changes (class 0, which
includes all the other changes).  This partitioning is based on
charge.  The list works on universal code.  For mt code (icode = 1),
some of the pairs specified cannot change between themselves in one
step and should be deleted.  I did not check carefully and the program
simply ignores such pairs.



5
1: DN HQ LI MI ML FI FL YF VM
2: QR ED EQ HR PA SN TP VI VL
3: GA KQ KE SG TA TN TS WL VA VF
4: HN HD PQ PH SP TK TM YH
0: all others

// end of file.
Partitioning according to grantham.dat, for table 5 in Yang et al. (1998).





Notes for file format: The number on the first line is the number of
substitution types or omega's.  This line is followed by as many lines
as the number of substitution types.  The last line, not read by the
program, includes all amino acid pairs not yet mentioned.  If the
specified amino acid pair cannot change between themselves in one step
(by one codon-position difference), the pair is discarded, with a
warning message.  To fit the "general" model, assigning an independent
w ratio for each one-step amino acid pair, put -1 at the start of the
file, and then the program will ignore the rest of the file and fit
the general model (see pages 1608-1609 in Yang et al. 1998).

Reference

Yang, Z., R. Nielsen and M. Hasegawa. 1998. Models of amino acid
substitution and applications to mitochondrial protein evolution. 
Mol. Biol. Evol. 15:1600-1611.