1. Introduction

Glaucoma is a complex, heterogeneous ocular disorder with multi factorial etiology characterized by structural damage to the optic nerve, and commonly associated with relatively high intraocular pressure (IOP) [1-2]. It is a leading cause of irreversible blindness worldwide with ~20% of cases occurring secondary to other ocular or systemic diseases [2-4].

Based on anatomical changes in the anterior chamber angle, primary glaucoma may be classified as primary angle closure glaucoma (PACG) or primary openangle glaucoma (POAG), which may be further subdivided into juvenile openangle glaucoma (JOAG) and adult onset POAG [1,5]. Glaucoma is a treatable disease if detected early; however, many patients are diagnosed during routine examinations or only following advanced field loss, as glaucoma is typically asymptomatic in the early stages. Therefore, the development of an accurate test for the detection of presymptomatic carriers at risk is important for the management of glaucoma.

A family history of glaucoma is a well-known risk factor and hence genetic background is considered an important factor for the development of the disease [6-8].

Several genes have been reported to be associated with primary glaucoma including myocilin (MYOC), WD repeat domain, neurotrophin 1, cytochrome P450 family 1 subtype [9-10]. To date, mutations in these genes account for only ~5% of patients with POAG, and the influence of mutations in these genes on patients with PACG remain controversial [11-12].

The MYOC gene, is located on chromosome 1q24.3q25.2. Mutations in the gene are commonly found in juvenile or early adult patients with high IOP although mutation frequencies vary between ethnic groups [13].

Bioinformatics is now playing a key role in different scientific areas. It involves computer sciences, mathematics, and statistics in order to analyze biological data that is being produced through the different sequencing techniques. Bio computing plays a key role in understanding the implication of genomic variations, especially single-nucleotide polymorphisms (SNPs), which represent the most frequent genetic variations in the human genome [14].

SNPs are the single base change in coding or non-coding DNA sequence and are present in every 200-300 bp in human genome [15]. The nonsynonymous SNPs (nsSNPs) are the single nucleotide variations that affect the coding region of the protein and modify the mutated site-encoded amino acid, which may lead to a structural modification of the mutated protein, and may thus cause function alteration [15].

The aim of the present study was to perform a computational analysis of the nsSNPs in the MYOC gene to identify the possible pathogenic SNPs and the effect they may impose on protein structure and function.

2. Materials and Methods

SNPs in human MYOC gene data wereobtained from The National Center for Biotechnology Information (NCBI) dbSNP database during February 2020. The data obtained was further analyzed using various software.

1- GeneMANIA

GeneMANIA (http://www.genemania.org) is a web interface that helps predicting the function of genes and gene sets, can be used to find new gene members of a pathway or complex. MYOC gene name was entered as an input for GeneMANIA and the results were shown as a diagram showing the genetic interactions, pathways, co-expression, co-localization and protein domain similarity [16].

2- Functional and structural analysis of SNPs

SNPs retrieved from the dbSNP database were analyzed according to the scheme shown in Fig.1.

Figure 1. Flow chart for SNP analysis

nsSNPs were analyzed using 7 prediction tools: SIFT, Polyphen-2, PROVEAN, SNPs & GO, PHD -SNP, I-Mutant 3.0 and project hope.

a. Sorting intolerant from tolerant (SIFT)

(http://siftdna.org/www/SIFT_dbSNP.html). It predicts the tolerated and deleterious SNPs and identifies the impact of amino acid substitution on protein function and phenotype alterations. It generates alignments with a large number of homologous sequences, and assigns scores to each residue ranging from zero to one. The input was the rs of the nsSNPs (obtained from the db SNP database) and the results were obtained as either deleterious or tolerated based on the score of 0.05 or less [17].

b. PROVEAN (Protein Variation Effect Analyzer)

(http://provean.jcvi.org/seq_submit.php). It is a software tool which predicts whether an amino acid substitution or indel has an impact on the biological function of a protein. The input was the protein sequence in FASTA format (obtained from Uniprot / Expasy-database) and the amino acid substitution [18].

c. Polymorphism Phenotyping (PolyPhen-2)

(http://genetics.bwh.harvard.edu/pph2/). It is an online bioinformatics program that automatically predicts the consequence of an amino acid change on the structure and function of a protein based on a number of features such as sequence, phylogenetic and structural information. The program searches for 3D protein structures, multiple alignments of homologous sequences and amino acid contact information in several protein structure databases, then calculates position-specific independent count scores (PSIC) for each of the two variants, and then computes the PSIC scores difference between two variants. The higher a PSIC score difference, the higher the functional impact a particular amino acid substitution is likely to have. The nsSNPs that were predicted to be intolerant by SIFT were submitted to Polyphen-2 as protein sequence in FASTA. Then the position for wild type and mutated amino acids were submitted. Prediction outcomes could be classified as benign, possibly damaging or probably damaging, according to the posterior probability intervals (0, 0.2), (0.2, 0.85) and (0.85, 1) respectively [19].

d. SNPs & GO and PHD-SNP

Predicting disease associated variations using GO terms (http://snps.biofold.org/phd-snp/phd-snp.html). SNPs & Go predicts whether the new phenotype derived from a ns SNP is disease related or not (neutral) [20]. The protein sequence was submitted to the program after providing position of the wild and the new amino acid residue. PHD-SNP also shows the same result and it is shown within the same program.

e. Effect of SNPs on Protein Stability

I-Mutant version 3.0 (http://gpcr2.biocomp.unibo.it/cgi/predictors/I-Mutant3.0/I-Mutant3.0.cgi) was used to predict the effect of the SNPs in the protein stability. I-Mutant basically can evaluate the stability change of a single site mutation starting from the protein structure or from the protein sequences [21]. The input was the protein sequence and position of wild and new amino acid residue. The output is classified into decrease or increase stability based on RI, and the DDG value.

3- Investigation of the structural effect

Project hope (HOPE; http://www. cmbi.ru.nl/hope/home) is an automatic mutant analysis server to study the insight structural features of native protein and the variant models. HOPE provides the 3D structural visualization of mutated proteins, and gives the results by using UniProt and DAS prediction servers. The input was the protein sequence and wild type and new amino acids. HOPE server predicts the output in the form of structural variation between mutant and wild type residues and the effect they have on protein structure and hence the function. [22].

3. Results

Fig.2 shows the co-expression. physical interaction, shared protein domain between various gene and MYOC gene network. Eight genes OLFM1, OLFM2 OLFM3, OLFM4 OLFML1, OLFML2A OLFML2B, OLFML3 were having a shared protein domain. (Appendix 1). These genes are parlogs for MYOC and hence the shared domains.

Figure 2. GeneMANIA result for MYOC gene

The total number of nsSNPs obtained from db SNPs was 109 SNPs. Using SIFT software 30 SNPs were found to be deleterious while 65 were found to be tolerated. Ten SNPs did not give any result using SIFT. These deleterious SNPs were further analyzed using Polyphen-2 which showed that 24 SNPs were probably damaging, 5 SNPs possibly damaging, while only one SNP were reported to be benign (Appendix 2). On the other hand PROVEAN results showed that 19 SNPs had a deleterious effect (Appendix 2). Only 17 SNPs were found to be disease related using both SNPs & GO and PHD- SNP (Appendix 2).

The total SNPs predicted to be disease related using the five different software were 16 SNPs as shown in Appendix 3.

Regarding the effect of the SNPs on protein stability, I- mutant results showed that 13 SNPs decreased the protein stability while 3 SNPs ( rs74315330, rs74315331 and

rs201573718) increased the protein activity

Project hope result for structural analysis:-

Four SNPs were investigated using project hope. They were selected based on polyphen 2 score (score =1) (Appendix 4).

4. Discussion

In this study, investigation of nsSNPs in the MYOC gene was done using different computational software. A total of 16 SNPs were reported to be damaging using five different soft wares.

A study carried among patients in Pakistan, showed the association of different SNPs in the MYOC gene and glaucoma, although with no statistical significance [23]. These SNPs (rs74315328, rs74315330, rs74315332, rs74315334, rs74315336, rs74315338 and rs121909193) were also confirmed in the current study to be disease related. Another SNP with rs74315341 has been reported among Caucasian and Brazilian population to be associated with glaucoma [24] but in this study this SNP was predicted to have a neutral effect using two software namely SNP and Go and PhD- SNP..

Two SNPs were also reported to be disease related in this study and were also detected among Australian population (rs74315330, and rs74315334 [25]. Another SNP rs rs74315329 was reported as an important risk factor among Tasmanian population [26], but it has not been predicted in the present study. Another two SNPs (rs74315328 and rs74315331) were reported to be associated with hereditary glaucoma in the United states [27]. and were also confirmed in this study.

Genetic defects can lead to an altered protein product which can be secreted into the extracellular matrix of the trabecular meshwork causing a severe form of autosomal dominant JOAG associated with very high IOP [9]. The effect of SNPs on protein structure can have different impacts such as increasing or decreasing its activity (as predicted by I Mutant) and hence affect the folding of the protein in the correct manner or affecting binding of the protein with specific types of ions or ligands as predicted by project hope- and can hence affect the function of the protein. Previous studies reported that mutated myocilin become tangled in the cell in its altered form [28].

5. Conclusions

The current study showed that 16 nsSNPs are associated with glaucoma using various computational tools. These mutations can distort the protein stability or it’s binding with other ligands and thus affecting its function.

Appendix 1: GeneMANIA Results of MYOC Gene

Appendix 1

Appendix 2: Result of SNP Analysis Using Various Software

Appendix 2

Appendix 3: Total Number of SNPs Predicted to be Disease Related Using Different Software

Appendix 3

Appendix 4: Project Hope Results

Appendix 4