Resolution: standard / high
Figure 9.
Predicted vs observed probabilities for the residue triple in HIV protease with the
largest 
in the PI2+ cohort (46-54-90). The black dots correspond to the best-fit independent model, while the red dots
correspond to the best-fit pair-term model (Equation 11). The solid line of slope
1 corresponds to perfect agreement of the predicted data with the observed, which
would be obtained if the three-body term were included.
in the PI2+ cohort (46-54-90). The black dots correspond to the best-fit independent model, while the red dots
correspond to the best-fit pair-term model (Equation 11). The solid line of slope
1 corresponds to perfect agreement of the predicted data with the observed, which
would be obtained if the three-body term were included.
Haq et al. BMC Bioinformatics 2009 10(Suppl 8):S10 doi:10.1186/1471-2105-10-S8-S10