Introduction

To tackle the problem of how to optimize the parameters of the PON predictors, this study used a variety of techniques.
(1) Investigated which of the MSA tools gave the best quality alignment by obtaining the best
       Total Column (TC) score from MSA benchmarks.
(2) Obtained control data consisting of known variations and protein sequences
(3) Investigated the best threshold for certain PON predictors and determine the best PON
        predictor using Matthews Correlation Coefficient (MCC) scores from the PON Bench-mark.
(4) Found the optimal MSAs for each tool and any generic optimal MSAs that would be used by
       all the PON tools.
(5) Stages one and three were automated due to the amount of data being processed.