Computational Predictions of Functional Single Nucleotide Polymorphisms in the Human Nerve Growth Gene

Abstract

Nerve Growth Factor (NGF) is a fundamental neurotrophin that plays a pivotal role in neuronal growth and survival, and is relevant to several diseases, including cancer. Single-nucleotide polymorphisms (SNPs), particularly non-synonymous SNPs (nsSNPs), can potentially alter protein function by changing its structure and stability. However, their impact on the NGF gene has been poorly characterized. We aimed to identify and characterize functionally impactful nsSNPs in NGF gene using computational tools. We retrieved nsSNPs from the NCBI dbSNP database and utilized various computational tools, including SIFT, PolyPhen-2, PredictSNP, PhD-SNP, PANTHER, and SNAP2, to predict their deleterious properties. I-Mutant 2.0, Project HOPE, Missense3D, and MutPred2 were utilized to assess the effects of mutations on protein stability and structure. The ConSurf server was used to determine the evolutionary conservation of high-risk residues. Besides, CScape, OncoVar, and canSAR Black databases predicted the oncogenic potential. Of the 268 nsSNPs analyzed, 36 were expected to be deleterious, with 32 contributing to NGF instability. Structural modeling showed drastic structural rearrangements of the critical residues, especially Q172R (rs1557933464) and Q172E (rs746593757), with enhanced flexibility and disruption of hydrogen bonding. ConSurf analysis revealed significant changes to the three-dimensional conformation of NGF, primarily in highly conserved regions, as determined by structural modeling using Phyre2 and SWISS-MODEL. We identified Q172E (rs746593757), E132K (rs772557857), V230A (rs767703003), and W142R (rs929155379) as putative oncogenic variants. All of these nsSNPs are within functional domains of the protein and may affect NGF-mediated signaling pathways. This study represents the first extensive in silico screening of deleterious nsSNPs within the NGF gene, suggesting their potential linkage to cancer. Our study highlights the utility of such computational predictions in prioritizing high-risk genetic variants; however, further experimental validation is needed to establish their role in disease pathogenesis and to assess their potential for therapeutic targeting.

J. of Sci. and Tech. Res. 6(2): 131-141, 2025