Use enhanced statistical model
The power of MSA comparison crucially depends on the quality of statistical models
used to rank the similarities found in a database search, so that biologically relevant
relationships are discriminated from spurious connections. We developed an accurate
statistical description of MSA comparison that does not originate from conventional
models of single sequence comparison and captures essential features of protein families. 
Specifically, we focus on (i) modeling local sequence patterns dictated by protein
secondary structure, and (ii) precise approximation of empirical statistical distributions
by a new type of analytical function, "power-EVD". 
This model surpasses conventional statistical models in the accuracy of detecting remote
protein similarities. 

 Ruslan I. Sadreyev and Nick V. Grishin (2008) Accurate statistical model of comparison
 between multiple sequence alignments. Nucleic Acids Res. 36(7):2240-8.