©1996-2019 All
Rights Reserved.
Online Journal of Bioinformatics. You may not store
these pages in any form except for your own personal use. All other usage or
distribution is illegal under international copyright treaties. Permission to use any of these pages in any other way besides the
before mentioned must be gained in writing from the publisher. This
article is exclusively copyrighted in its entirety to OJB publications. This
article may be copied once but may not be reproduced or re-transmitted without
the express permission of the editors.
OJB©
Online Journal of Bioinformatics©
Volume 6 : 42-50, 2005.
Substitution of
polymorphic amino acids in proteins:Predicting
functional alterations using amino acid properties.
Baksi K, Bhattacharyya NP.
Crystallography and Molecular Biology Division, Saha Institute of Nuclear Physics, Kolkata 700 064, India.
ABSTRACT
Baksi K., Bhattacharyya NP. Substitution of polymorphic amino acid in proteins:Predicting functional
alterations using amino acid properties. Onl J Bioinform., 6:42-50, 2005. Polymorphic
amino acid substitutions (AAS) among non-synonymous single nucleotide
polymorphisms (nsSNPs), likely to alter the function
of proteins were prioritized. Calculated changes in molecular volume (δV), molecular mass (δM),
molecular surface (δS) and polarity (δP) of amino acids in 1018 evolutionary neutral (set
I) AASs and 566 AASs that are known to alter the function (set II)
(training sets) were calculated. Mean values of δV,
δM, δS and δP were (28.5 ± 21.8) Å3,
(23.6 ± 18.1) D, (34.4 ±25.2) Å2 and (1.3 ± 1.4) units respectively
for set I. For set II, these values were (48.4 ± 35.0) Å3, (36.8 ±
24.0) D, (55.3 ± 35.7) Å2 and (2.9 ± 2.0) units respectively which
were higher than that obtained in the neutral data set I (for all of the
parameters p < 0.0001, t test). Linear Discriminant Analysis (LDA) revealed
that both sets had significant differences between their centroid positions for
all four parameters taken together as variables (Wilks’ Lambda=0.756,
p<0.001). The findings showed that depending on the changes in the above
properties of the amino acids, the training sets could be differentiated. On the
basis of this observation, we proposed that changes in any one of the
parameters beyond δV > 50Å3, δM > 40D, δS> 60
Å2 and δP
> 3 units were likely to alter the function of the proteins. Using
this prediction rule, we analyzed 756 AAS in an independent evolutionary data
set (IA), 15,066 AAS from Human Mutation Database (HGMD, set III) and 3,747
validated nsSNPs AAS (set IV). We predicted that 30%
of AAS in set IA, 69% AAS in set III and 44% AAS of set IV were likely to alter
the function of the proteins by such substitutions. False positive error rate
for this prediction was thus 30%. Subtracting the possible false prediction
data (30%), about 14% of the validated nsSNPs is
likely to be deleterious. We compared this prediction rule with BLOSUM62
and SIFT, and obtained comparable results. The findings suggest that the simple
amino acid based prediction (SABP) rule, based on the intrinsic properties of
the amino acids only, could be used to prioritize nsSNPs
and used for association studies in complex genetic diseases.
KEYWORDS: Amino Acid,
protein, prediction