Unraveling genetic signatures: the impact of SNPs on therapy response in inflammatory bowel disease patients Authors: Nesa Kazemifard, Masomeh Askari, Leila Navapour, Anahita Tajik, Mina Faraji, Arvand Akbari, Navid Mogharrab, Shaghayegh Baradaran Ghavami, Hamid Asadzadeh Aghdaei Published: 2025 ABSTRACT Background:Inflammatory bowel disease (IBD) is a chronic condition characterized by immune dysregulation and intestinal inflammation, with anti-tumor necrosis factor α (anti-TNFα) therapies being a key treatment strategy. However, therapeutic responses vary across patients due to genetic factors, necessitating the identification of predictive biomarkers to optimize outcomes. This study employs bioinformatic approaches to investigate the deleterious effects of SNPs in IBD-associated genes, focusing on the structural implications of HFE gene variants and the impact of TNFRSF1B 3' UTR SNPs on miRNA binding site alterations.Method:The study utilized a systematic literature review to identify genes influencing drug response in IBD, followed by SNP detection. SNPs were filtered based on population frequencies using global databases (gnomAD, ExAC) and Iranome for region-specific data. Structural and functional impacts of missense variants were evaluated using in silico tools and protein domain analysis via InterPro and UniProt. Additionally, molecular dynamics simulation indicated structural changes in HFE gene variants, RNA-seq data analyzed differential gene expression, and miRNASNP-v3 predicted miRNA binding site alterations in 3' UTR regions. Result:A comprehensive bioinformatics approach identified 41 key IBD drug response genes contributing to cellular pathways, with 69 coding sequence SNPs and 27 3'-UTR SNPs selected for analysis. Coding SNPs, including ABCB1 and HFE variants, were predicted to significantly alter protein structure and stability. RNA-seq analysis revealed TNFRSF1B upregulation in IBD patients, and its 3'-UTR SNPs were shown to affect miRNA binding. Molecular dynamics simulation demonstrated structural instability in the HFE H63D variant, suggesting functional disruption. Conclusion:This study explores genetic variations influencing anti-TNF therapy responses in IBD patients. Polymorphisms like HFE rs1799945 and TNFRSF1B SNPs are identified as pivotal markers for personalized treatment strategies. Computational findings highlight their clinical potential, emphasizing the need for experimental validation to advance precision medicine and improve therapeutic outcomes and quality of life in IBD management. Introduction Inflammatory bowel disease (IBD) is a chronic and recurrent inflammatory disorder of the gastrointestinal tract, primarily comprising two types: ulcerative colitis (UC) and Crohn's disease (CD). With its rising incidence in recent years, IBD is now considered a global condition that imposes significant financial burdens on individuals, families, and society (1). Chronic gut inflammation can be driven by genetic, environmental, and host–microbiome interactions (2). To gain further insights into IBD’s genetic susceptibility, twin studies have been conducted, showing increased concordance for both CD and UC, including a 58% concordance among monozygotic twins with CD. Moreover, first-degree relatives of IBD patients have a five-fold higher risk of developing the disease (3). Genome-wide association studies (GWAS) have identified 41 CD-specific loci, 30 UC-specific loci, and 137 shared loci. Interestingly, approximately 90% of these loci are located in noncoding regions (4). Currently, no definitive cure for IBD exists. Therefore, treatment strategies focus on controlling symptoms and maintaining remission, defined as complete mucosal healing and the absence of clinical symptoms. However, achieving optimal therapy remains challenging due to heterogeneity in the clinical presentation and genetic background of UC and CD (5). Drug therapies for IBD generally follow two therapeutic principles: (i) blocking cytokine signaling and (ii) inhibiting immune cell trafficking. Traditional therapies (such as steroids, azathioprine, methotrexate, and mesalamine) act through immunosuppressive, anti-metabolic, or unclear mechanisms and are often used as first-line treatments (6). Next-generation therapies include antibody-based biological agents, such as anti–tumor necrosis factor (TNF) antibodies (Infliximab, Adalimumab), anti–interleukin (IL)-12/23 receptor antibodies (Ustekinumab) (7), and anti-integrin α₄β₇ antibodies (Vedolizumab) (8). More recently, small-molecule agents, including the Janus kinase (JAK) inhibitor Tofacitinib, have also been introduced into clinical practice (9, 10). Despite their success, these therapies are not universally effective. For instance, approximately one-third of patients treated with anti-TNF agents exhibit a primary non-response, while 30–50% of initial responders eventually experience secondary loss of response after multiple treatment courses (11). Specifically, around 13% of patients lose response to Infliximab each year (12). Thus, reliable biomarkers that can predict therapeutic success are urgently needed to reduce costs, minimize adverse effects, and enable personalized medicine approaches that optimize drug selection for individual patients (13). Given that drug response variability is strongly influenced by genetic factors, single nucleotide polymorphisms (SNPs) in genes involved in immune regulation and inflammation represent promising biomarkers for predicting therapeutic outcomes (14, 15). Although numerous studies have investigated the role of SNP biomarkers in IBD therapy response, a critical gap remains: the lack of a comprehensive bioinformatics framework to systematically assess the deleterious potential of SNPs within drug response–related genes. To address this, the present study employs bioinformatics tools to evaluate exonic and 3′ untranslated region (UTR) SNPs in therapy-related genes, with the aim of identifying potential biomarkers for predicting treatment response in IBD patients. Methods Identification of candidate variants involved in Drug response in IBD A systematic literature review was conducted to identify genes that contribute to the drug response in patients with IBD (15). Additionally, the TNF receptor expressed in the colon was selected for further investigation based on its role in drug response to anti-TNFα monoclonal antibodies in IBD patients. The biomaRt package in the R programming language was used to determine the chromosomal locations of SNPs within a gene of interest. The annotations were based on the GRch38 reference genome, which provides accurate and current genomic information. To find SNPs in our candidate gene regions, the tabix function in the samtools program was applied. Related pathways for each gene were also obtained using the PubChem database (www.pubchem.ncbi.nlm.nih.gov). Filtration of SNPs based on their frequencies in population SNPs were filtered based on their reported global frequency in the 1000 genome (www.internationalgenome.org), the Genome Aggregation Database (gnomAD) (www.gnomad.broadinstitute.org) and The Exome Aggregation Consortium (ExAC) databases (http://exac.broadinstitute.org) Which provide assessment of sequence-resolved structural variants diverse global populations and specific populations. SNPs with a global frequency greater than 0.05 were selected. Also, Iranome database (www.iranome.ir) were used for filtering SNPs that are specifically have high frequencies in Iranian population. Assessment of structural and functional consequences of missense variants using in silico tools Located in coding regions, nonsynonymous or missense SNPs lead to variation in protein products of genes resulting from amino acid substitution. In order to predict the structural and functional influences of missense SNPs on human proteins, various in silico tools were used. The Polymorphism Phenotyping v2 (PolyPhen-2) (http://genetics.bwh. harvard.edu/pph2/) database is able to evaluate the impact of amino acid substitutions on function and structure of proteins based on sequence, phylogenetic as well as structure-based features. Amino acid allelic variants are estimated qualitatively, as benign (scores from 0.0 to 0.15), possibly damaging (scores from 0.15 to 0.85), or probably damaging (scores from 0.85 to 1.0), respectively (16). Another web-based resource, Protein Variation Effect Analyzer (PROVEAN) (http://provean.jcvi. org/index.php), is a high accurate tool that provides functional predictions for single or multi amino acid substitutions, insertions, and deletions for both human and mouse. The PROVEAN predictions are classified in two groups based on the defaulted cutoff score (-2.5): deleterious variant with score < − 2.5 and neutral variant with score > − 2 (17). The third database, Sorting Intolerant From Tolerant (SIFT) (https://sift.bii.a-star.edu. sg/) is capable of predicting whether an amino acid replacement will result in protein functional change and alter phenotype, possibly. This tool differentiates functionally neutral from deleterious amino acid substitutions based on the quantitative scores ranging from 0 to 0.05 and 0.05 to 1, respectively (18). Functional Analysis through Hidden Markov Models (FATHMM) constitutes a valuable tool for the prediction of the functional impact of protein missense mutations. This tool employs 'pathogenicity weights' to represent the overall tolerance of the protein/domain to mutations. The generated score spans from -16.13 to 10.64, with lower scores indicating a higher probability of being deleterious. Scores below -1.5 are classified as deleterious (19). Domains are functional and critical part of protein structure and any changes in their amino acid sequence could impact on their role. The InterPro (www.ebi.ac.uk/interpro/) and UniProt (www.uniprot.org) online databases were searched to determine whether the SNP is present in a protein domain sequence. Additionally, the post-translational modifications (PTMs) of each altered amino acid in the protein structure were examined using the UniProt database. Evaluation of missense variants effect on protein stability I-Mutant2.0, a support vector machine (SVM)-based online tool (https://folding.biofold.org/cgi-bin/i-mutant2.0.cgi) were used to predict non-synonymous SNPs effect of protein structure and stability. This tool uses the ProTherm data and its prediction is based on free energy changes of protein stability upon single amino acid changes under different conditions (20). Based on the results of this data base, a decrease (I mutant value < 0) or an increase (I mutant value > 0) stability were reported for each non-synonymous SNPs. Higher I mutant value shows higher impact of polymorphism on protein free energy and stability. Exploring structural dynamics of HFE gene variant The molecular dynamics (MD) simulations of the wild type (WT) HFE and H63D variant structures were performed by GROMACS package version 5.0.5 (21) using the CHARMM36 force field (22). First, the atomic coordinates of WT protein were retrieved from Protein Data Bank (PDB ID: 1DE4) (23) and three-dimensional structure of the variant protein was modeled using Phyre2 web server (24). Each protein was placed in a cubic box with at least 10 Å distance to the edges and solvated with TIP3P water molecules. To neutralize the system charge, an appropriate number of water molecules was replaced by counter ions. Then, both systems were subjected to energy minimization and equilibration followed by a 100 nanoseconds (ns) MD simulation as previously detailed in (25). After MD simulation, the trajectories were analyzed using built-in functions of the GROMACS package version 5.0.5. The three-dimensional structures were visualized by UCSF Chimera version 1.17.1 (26). Analysis of RNA-seq data The RNA-seq raw data counts were downloaded from the GEO repository (www.ncbi.nlm.nih.gov/geo) with the accession number GSE83687 including samples of 134 patients undergoing bowel resection. Among these samples, colonic tissue samples were used (N=89) containing 42 IBD and 47 controls. Control samples were harvested more than 10 cm away from the tumor and normal non inflamed condition from patients undergoing bowel resection. Samples were sequenced by Illumina HiSeq 2500 (GPL16791) (27). Log2 transformation and normalization were done for count data. For obtaining differentially expressed genes (DEGs) between IBD and control samples, the DEseq2 package was used in R and principal component analysis (PCA) plot was drawn by ggplot2 package. For filtering significant DEGs, adjusted p-value <0.05 and |logFC| > 1 were selected as the threshold. Determination if miRNA target for 3’UTR variations For predicting the 3’UTR regions SNPs effect on drug response, the miRNASNP-v3 online tool (http://bioinfo.life.hust.edu.cn/miRNASNP#!/snp_summary) were used to explore the impact of the SNP on mRNA:miRNA interaction. The results showed several gain and loss of targets for miRNAs in 3’UTR SNP regions. Table 1Cellular pathways for each selected gene involve in the anti-TNF therapy responsePathwayDatabaseGeneToll-like receptor signaling pathwayWikipathways Ly96 CD40 Cell Death and divisionReactome &Gene onthology CASP9 NGFR FAS CCNY RNF175 TransportersReactome HFE ABCB1 TAP1 AutophagyReactome ATG16L1 NOD2 Prostaglandin synthesis and regulationWikipathways PTGER4 DNA repairGene onthology EMSY (C11orf30) Cytokine Signaling in Immune systemReactome TNFRSF4 TNFRSF6B TNFRSF10A TNFRSF10B TNFRSF10C TNFRSF10D TNFRSF11A TNFRSF11B TNFRSF12A TNFRSF13B TNFRSF13D TNFRSF14 TNFRSF17 TNFRSF18 TNFRSF19 TNFRSF21 TNFRSF25 EDA2R IFNG IL1R LTBR IL13 IL1RN IL17A TNFAIP3 IL23R IL27 Cellular pathways for each selected gene involve in the anti-TNF therapy response Table 2Deleterious SNP variants based on SIFT, Poly-Phen, FATHMM and PROVEN prediction models. I mutant scores are also presented for predicting the SNP impact on protein stability. SIFT scoring system: Deleterious (0 to 0.05) and neutral (0.05 to 1). Poly-Phen scoring system: Benign (0.0 to 0.15), possibly damaging (0.15 to 0.85), or probably damaging (0.85 to 1.0). FATHMM scoring system: Deleterious (score < -1.5) and neutral (score > -1.5). PROVEN scoring system: Deleterious (score < − 2.5) and neutral (score > − 2). I mutant scoring system: A decrease in protein stability (I mutant score < 0) or an increase in protein stability (I mutant value > 0). Deleterious and damaging scores are marked by *.Gene NameVariant CodeChangeClinVar significanceSIFT ScorePoly-Phen ScoreFATHMM ScorePROVEN ScoreI mutant Score ABCB1 rs1045642A>GBenign/drug response0.001*0.915*-2.82*-2.78*-1.41 IL23R rs1884444G>T-0.3810.002-1.97*0.05-0.56 IL23R rs11209026G>AInflammatory Bowel Disease, Psoriasis0*1*0.71-1.16-1.14 IL27 rs181206A>G-0.002*0.99*1.13-2.95*-1.12 NGFR rs2072446C>G-0.1090.008-0.25-1.90.18 NOD2 rs5743291G>ASarcoidosis / Chron’s Disease / Blau Syndrome0.027*0.51*0.65-0.44-0.47 RNF175 rs1337C>G-0.1170.985*1.54*-2.32-0.98 RNF175 rs10517577T>A-0.120.002-1.75*-1.7-0.76 TNFRSF10A rs20576T>G-0.0530.712*-1.45-2.1-1.25 TNFRSF10B rs1047266G>A-11*-1.2-2.30.12 TNFRSF11A rs35211496C>TBone Paget disease / Osteoporosis0.044*0.1400.07-2.65*0.65 TAP1 rs1135216T>C-10-2.37*5.78-1 HFE rs1799945C>GPathogenic0.016*0.142-2.42*-1.23-0.38 TNFAIP3 rs2230926T>G-0.2210.0081.52-4.87*-1.54 Deleterious SNP variants based on SIFT, Poly-Phen, FATHMM and PROVEN prediction models. I mutant scores are also presented for predicting the SNP impact on protein stability. SIFT scoring system: Deleterious (0 to 0.05) and neutral (0.05 to 1). Poly-Phen scoring system: Benign (0.0 to 0.15), possibly damaging (0.15 to 0.85), or probably damaging (0.85 to 1.0). FATHMM scoring system: Deleterious (score < -1.5) and neutral (score > -1.5). PROVEN scoring system: Deleterious (score < − 2.5) and neutral (score > − 2). I mutant scoring system: A decrease in protein stability (I mutant score < 0) or an increase in protein stability (I mutant value > 0). Deleterious and damaging scores are marked by *. Result Drug response genes are involving in several cellular pathways Searching literature review, genes playing significant roles in the drug response in IBD were gained and investigated in HGNC database and revised, accordingly. As a consequence, 41 key genes involved in drug response were selected. Extracted genes contributed to various signaling pathways in drug response process. The biological roles of above-mentioned genes with specific focus on drug response were summarized in Table 1. In the following, genetic variations in different regions of the corresponding genes, namely coding sequence and 3'-UTR, were obtained from the dbSNP. Accordingly, 31,097 SNP in the coding sequence as well as 20,318 SNP in 3'-UTR region were gained, respectively. After filtering variants based on their global frequency (Grater than 0.05), 69 exonic variants (35 synonymous and 34 non-synonymous mutations) and 27 3’UTR SNPs remain for further analysis. SNPs in the coding region potentially modify protein structure and function The effect of coding region SNPs on drug response were examined using 4 common web-based resources, SIFT, PolyPhen-2, PROVEAN and I-mutant 2.0. Our results predicted that the structures of candidate genes are under the influence of missense SNPs, potentially resulting in protein biological function alterations (Table 2). Based on SIFT databases, the number of 6 SNPs (ABCB1, IL23R, IL27, NOD2, TNFRSF11A, HFE) were predicted as damaging and deleterious. Regarding Polyphen-2 database, 7 SNPs in ABCB1, IL23R, IL27, NOD2, RNF175 and TNFRSF10A/B genes were predicted as possibly or probably damaging. Conversely, the FATHMM and PROVEN databases identified 6 and 4 SNPs as deleterious, respectively. Specifically, the FATHMM database predicted deleterious effects for SNPs in ABCB1, IL23R, RNF175, TAP1, and HFE, whereas the PROVEN database identified SNPs in ABCB1, IL27, TNFRSF11A, and TNFAIP3 as deleterious. These findings underscore the utility of computational tools in predicting the functional impact of genetic variants. Moreover, according to I-mutant 2.0, the stability of 3 proteins increased while that of 11 proteins decreased following occurrence of SNPs in the coding sequence regions in probably damaging SNP candidates. Furthermore, the ABCB1 rs1045642 variant has been consistently predicted to exert deleterious effects across all predictive databases. It can be noticed that some protein structure changes resulting from SNPs were confirmed in two databases, simultaneously. Among probably damaging SNP variants, the HFE gene (rs1799945) as a pathogenic variant from two prediction algorithms (SIFT & FATHMM) and ClinVar with available 3D structure were selected for further structural analysis. TNFRSF1B is upregulated in IBD patients compare to controls After conducting RNA-seq analysis of GSE83687 data, we identified 2071 significantly up-regulated genes and 1055 down-regulated genes in IBD patients. We filtered the dysregulated genes based on our anti-TNF therapy response genes and found that only TNFRSF1B was significantly up-regulated in IBD patients, with a 1.106-fold increase and an adjusted p-value of 1.728e-09. SNPs in the 3'-UTR region potentially regulate drug response through miRNA binding efficiency alteration and changes in TNFRSF1B expression The potential changes imposed by the 3′-UTR SNPs on miRNA binding proficiency in drug response related genes were assessed using miRNASNP-v3 (http://bioinfo.life.hust.edu.cn/miRNASNP/). according to the RNA-seq analysis results, we assessed the influence of 3'UTR SNPs of TNFRSF1B on the interaction of miRNAs. Our findings suggest that when the minor allele frequency (MAF) is greater than or equal to 0.1, five SNPs in the 3'-UTR region of the TNFRSF1B gene may lead to the introduction of miRNA target sequences and a change in protein expression. Specifically, the SNPs rs1061624, rs3397, rs5746065, rs1061628, and rs1061631 in the 3’UTR of the TNFRSF1B gene are predicted to cause 23 miRNAs to lose their interaction site and 16 miRNAs to gain binding sites on the 3’UTR of the gene, potentially influencing its expression (Table 3). HFE H63D variant shows structural and dynamical changes that may influence protein function We performed 100-ns molecular dynamics simulations for WT and H63D variant of HFE protein (rs1799945). In order to explore the structural changes of the proteins, the root-mean-square deviation (RMSD) and radius of gyration (Rg) were analyzed along the MD simulation time. The RMSD analysis showed that WT protein reached a stable conformation approximately after 50 ns. Despite that, the H63D variant showed high variation in the position of alpha carbon (Cα) atoms (Figure 1A) and a similar trend in Rg (Figure 1B), which implies that the variant protein experienced high conformational changes upon substitution. The three-dimensional structures extracted every 10 ns from the last 50 ns of the trajectories confirm high structural instability of H63D variant (Figure 2) The principal component analysis (PCA) during the last 50 ns of the simulation confirmed a significant increase in the overall flexibility of the protein (Figure 3). Root mean square deviation (RMSD) analysis. (A) Time dependence of the RMSD of Cα atoms. The RMSD analysis showed that wild type (WT) protein reached a stable conformation approximately after 50 ns. Despite that, the H63D variant showed high variation in the position of alpha carbon (Cα) atoms. (B) Radius of gyration (Rg) during 100 ns MD simulation. The variant protein experienced high conformational changes upon substitution Three-dimensional structure analysis. Snapshots extracted every 10 ns from the last 50 ns of the (A) WT and (B) H63D trajectories confirms high structural instability of H63D variant. Principal component analysis (PCA). The projection of protein motion in phase space along the first two principal eigenvectors during the final 50 ns of the simulation reveals a marked increase in overall flexibility in the H63D variant compared to the Wild Type (WT). Table 2miRNA target changes due to various SNPs in 3'UTR of TNFRSF1B gene retrieved from miRNASNP-v3 database.rs codemiRNA nameFunctionrs1061624hsa-miR-639Loss of functionrs3397hsa-miR-4462hsa-miR-122-3phsa-miR-642b-3pGain of functionhsa-miR-362-3phsa-miR-329-3phsa-miR-7153-3phsa-miR-1228-3phsa-miR-603hsa-miR-8485hsa-miR-642a-3prs5746065hsa-miR-2467-3pLoss of functionhsa-miR-6721-5phsa-miR-637hsa-miR-1227-5phsa-miR-7155-5phsa-miR-486-3phsa-miR-8089hsa-miR-4700-5phsa-miR-4667-5phsa-miR-10393-3pGain of functionhsa-miR-6861-5phsa-miR-6801-5phsa-miR-6776-5prs1061628hsa-miR-4715-5pLoss of functionhsa-miR-4259hsa-miR-96-5phsa-miR-382-5phsa-miR-182-5phsa-miR-4642hsa-miR-1256hsa-miR-495-5phsa-miR-1271-5phsa-miR-12133hsa-miR-374b-3pGain of functionhsa-miR-1179rs1061631hsa-miR-662Loss of functionhsa-miR-660-5pGain of functionhsa-miR-425-5p miRNA target changes due to various SNPs in 3'UTR of TNFRSF1B gene retrieved from miRNASNP-v3 database. Solvent accessible surface area (SASA) and time-dependent secondary structure analysis (A) SASA increased in H63D variant, (B, C) Number of α-helix and β-sheet forming residues during the last 50 ns of the simulation decreased in H63D variant compared to wild type (WT) (Box plots for WT and H63D are shown in white and red, respectively). The trace value of the diagonalized covariance matrix calculated for the Cα atomic positional fluctuations was increased from 3.5 nm2 in WT to 79.5 nm2 in H63D variant. It was also observed that the solvent accessible surface area (SASA) increased in H63D variant (Figure 4A). On the other hand, time-dependent secondary structure analysis showed that α-helical and β-sheet content was decreased upon substitution (Figure 4B and 4C). It should be noted that no α-helix or β-sheet structures were completely lost, just their length were reduced. Although the average number of hydrogen bonds showed a little difference between the WT (219 ± 6) and H63D variant (215 ± 7), the number of hydrogen bonds with existence percentage greater than 70% was decreased from 167 in WT to 150 in variant protein. More detailed analysis revealed a strong hydrogen bond (94% occupancy) between Gln233 and Gln56 (Figure 5) which is lost in the variant protein. It seems that the structural instability of the H63D variant is due to the loss of Gln233-Gln56 hydrogen bond Discussion IBD is a chronic inflammatory disorder of the digestive tract. One of the most widely used treatment strategies for IBD has been the development of anti–tumor necrosis factor-α (anti-TNFα) agents, which block the pro-inflammatory cytokine TNF-α (28). However, anti-TNF therapy is not effective in all patients; approximately 20% are primary non-responders, while more than 30% experience a secondary loss of response (29). Thus, identifying both the underlying causes and predictive biomarkers of therapeutic response is crucial prior to determining treatment strategies. Our previous systematic review identified genes associated with treatment response in IBD patients (15). Building on this, the objective of the present study was to perform a comprehensive bioinformatics analysis of SNPs to identify potential predictive markers of anti-TNF response. This analysis was conducted using genes identified in the systematic review. For each candidate gene, cellular pathways were mapped, and all recorded SNPs were retrieved. Multiple prediction models were employed to assess their potential functional impact. Several SNPs were predicted to be deleterious, with significant consequences for protein structure and function. The Hydrogen bond between Gln233 and Gln56 in wild type (WT) protein which is lost in the H63D variant protein. The dashed green line denotes the hydrogen bond. Through the analysis of 41 candidate genes, it was found that several pathways are involved in the response to anti-TNF therapy. These pathways include Toll-like receptor, cell death and division, transport of small molecules, autophagy, prostaglandin synthesis and regulation, DNA repair, and cytokine signaling. Analysis of 41 candidate genes revealed that multiple pathways are involved in anti-TNF therapy response, including Toll-like receptor (TLR) signaling, cell death and division, small-molecule transport, autophagy, prostaglandin synthesis and regulation, DNA repair, and cytokine signaling. Toll-like receptors (TLRs) recognize pathogen-associated molecular patterns (PAMPs) on the surface of cells. Upon recognition, these receptors become dimerized and trigger downstream signaling cascades that cause inflammation, proliferation, and differentiation of Th1 and Th2 T cells, as well as dendritic cell maturation (30). Furthermore, genetic variation in TLR signaling genes shown to be associated with anti-TNF therapy response in rheumatoid arteritis patients (31). Another study reported that anti-TNF treatment response may be influenced by genetic variability in cell migration and viability pathways such as FAS and CASP9 (32) as we reported cell death and division pathways for our candidate genes. NGFR rs2072446 variant were reported as a probable variant related to response to the therapy in IBD patients. P75 neurotrophin receptor (encoded by the NGFR gene) is a low-affinity neurotrophins receptor, and involves in neuronal growth, apoptosis, and synapse plasticity. The rs2072446 SNP previously reported to be associated in amyloid-β deposition in Alzheimer disease (AD) and Psychiatric Disorders (33,34). However, Nerve growth factor also has a potential role in the maintenance of enteric nerves and may be associated with IBD. A study reported that UC patients have significantly higher p75 levels in lamina propria cells (35). Other possibly deleterious SNPs in this study, includes NOD2 rs5743291 polymorphism. NOD2 is among the most extensively studied genes associated with CD and plays a pivotal role in maintaining intestinal barrier integrity and immune homeostasis. It is critically involved in ensuring proper autophagic processes and regulating the composition of the gut microbiota. Deficiencies in NOD2 function have been linked to an increased susceptibility to CD, highlighting its significance in disease pathogenesis. Furthermore, recent studies have proposed NOD2 as a promising therapeutic target for CD, offering potential avenues for the development of targeted interventions to address underlying molecular mechanisms (36). NOD2 polymorphisms have been widely investigated across various populations due to their potential role in IBD. However, the rs5743291 polymorphism has been specifically analyzed in a single study involving the Indian population. This study concluded that no significant association was found between the rs5743291 polymorphism and the development of CD or UC (37). These findings suggest that the role of this particular polymorphism in disease susceptibility may vary across different populations and warrants further investigation to clarify its impact. Based on our study, the ABCB1 rs1045642 polymorphism has been identified as deleterious. The ABCB1 gene, also known as MDR1, encodes a transmembrane efflux pump that plays a crucial role in drug disposition and response by actively transporting a wide range of xenobiotics out of cells. This efflux mechanism is particularly significant in physiological barriers, such as the gastrointestinal tract, where it contributes to the processes of drug absorption, distribution, metabolism, and excretion. Notably, the C allele of the rs1045642 polymorphism has been reported to exhibit a protective role in the development of UC. Furthermore, this SNP has been associated with the therapeutic response to various drugs, including glucocorticoids, highlighting its relevance in personalized medicine and treatment strategies (38,39). Among the studied SNPs, rs1799945 of HFE gene predicted to be a non-synonymous polymorphism which can shift protein conformation and decreased stability that is related to loss of anti-TNF therapy response in IBD patients. Repnik et al previously reported that there is a strong association between rs2071303 SNP in HFE gene and response to adalimumab treatment in IBD patients (40). The HFE gene is responsible for the manifestation of hereditary hemochromatosis, a commonly encountered variety of iron overload disease (41). Its potential relevance in the context of IBD can be ascertained that patients with active Crohn's disease experience a compromised ability to absorb iron orally. The transmembrane glycoprotein generated by the HFE gene plays a key role in regulating iron uptake, thereby making it a potential subject of interest for IBD investigations (40). Iron metabolism plays a crucial role in various physiological processes, including cellular oxidative metabolism, immune system responses, and host defense mechanisms. It restricts iron availability to impede the growth of parasites or bacteria, thereby contributing to the body's defense system (42). In patients with HFE-associated hemochromatosis (HH), elevated levels of interleukin 8 have been reported. Specifically, in the H63D variant (rs1799945), secretion of matrix metalloproteinase 8 (MMP-8) and other neutrophil-secreted proteins was higher compared to controls (43). On the other hand, CD8 T lymphocytes could down regulate the HFE expression trough TNF and IFN-γ secretion (44). The HFE gene is also associated with TNF secretion for iron uptake. In an experimental study, it was observed that HFE knockout mice did not exhibit an increase in TNF expression following exposure to an iron-enriched diet, in contrast to the response observed in normal controls (45). Our study elucidates the structural and functional implications of the rs1799945 variant in the HFE gene, demonstrating its potential impact on therapeutic outcomes in patients with IBD undergoing anti-TNF therapy. The rs1799945 mutation precipitates notable alterations in the three-dimensional configuration of the HFE protein, specifically resulting in decreased α-helical and β-sheet content. Concurrently, this mutation augments the overall flexibility of the protein and significantly increases its solvent-accessible surface area (SASA). A critical structural disruption observed in the mutant protein is the loss of a strong hydrogen bond between Gln233 and Gln56, which is hypothesized to be pivotal for the protein's stability and function. These structural changes are posited to affect the protein’s functionality, potentially influencing the biological pathways involved in the inflammatory response and thus modulating the efficacy of anti-TNF therapy in IBD patients. TNF-α binds to its cell surface receptors: p55 TNF-α receptor (TNFRI) encoded by the TNFRSF1A gene and p75 receptor (TNFRII) encoded by the TNFRSF1B gene, initiating its inflammatory cascade (46). According to published documents, the TNFRSF1B gene may affect the likelihood of developing auto-inflammatory conditions like IBD and rheumatoid arthritis (RA). Additionally, TNFRSF1B variations could potentially predict how a person will respond to biological drugs. It has been reported that the initial response to anti-TNF therapy in CD patients can be predicted by the TNFRSF1B rs5746053 polymorphism (47). In a recent study, rs1061622 and rs3397 in TNFRSF1B found to correlate with differences in steroid-free remission rates and therapy persistence in IBD patients, suggesting their potential utility as predictive biomarkers for personalized treatment strategies (48). Moreover, variations in the TNFRSF1B gene are significantly associated with the presence of anti-neutrophil cytoplasmic antibodies (ANCA) in individuals with IBD (49). In psoriasis and RA patients also reported that there is an association between TNFRSF1B polymorphisms and positive response to anti-TNF therapy (50,51). The response to anti-TNF therapy agents is associated with changes in TNFRSF1B expression too. A study found that TNFRSF1B is significantly downregulated in non-responder CD patients (52). TNFRSF1B rs1061622 polymorphism was also reported to be linked to the long-term efficacy of Etanercept for treating patients with ankylosing spondylitis (53). Based on our study, TNFRSF1B SNPs have the potential to induce the loss or gain of miRNA binding sites within the 3' UTR of the mRNA. These alterations may significantly impact the expression of TNFRSF1B in patients with IBD. Consequently, such polymorphisms could lead to epigenetic modifications in gene expression, potentially influencing susceptibility to IBD or modulating the therapeutic response to anti-TNF treatments. These findings underscore the importance of further investigation into the epigenetic and molecular mechanisms underlying the role of TNFRSF1B SNPs in IBD pathogenesis and therapy. Several biomarker panels have been developed to predict therapeutic response in patients with IBD. For example, a recent study by Ghavami et al. introduced a microbiome-based panel capable of predicting responsiveness to Adalimumab therapy. In responders, stool microbiota profiles were characterized by increased abundances of Bacteroidetes and reduced levels of Proteobacteria, alongside enrichment of beneficial taxa such as Faecalibacterium prausnitzii, Bifidobacterium, and Akkermansia muciniphila (54). In parallel, transcriptome-based approaches have also been explored. Notably, Rac1/pSTAT3 expression in T cells has emerged as a thiopurine-specific pharmacodynamic (PD) biomarker panel, demonstrating superior accuracy compared to conventional pharmacokinetic therapeutic drug monitoring (TDM) in evaluating treatment efficacy (55). Furthermore, Joustra et al. recently reported that peripheral blood DNA methylation signatures can reliably predict patient response to vedolizumab and ustekinumab in Crohn’s disease, although with lower predictive value for Adalimumab. These methylation-based markers were shown to outperform standard clinical tools, thereby advancing precision medicine approaches and reducing reliance on trial-and-error strategies (56). In our study, we propose the use of SNPs in therapy-response–related genes as a novel panel, which, following experimental validation, holds potential for guiding drug response prediction and supporting personalized treatment strategies. Collectively, evidence suggests that an integrative multi-omics approach, combining genomic, transcriptomic, methylomic, and metagenomic panels, offers the most accurate framework for predicting treatment outcomes in IBD. Bioinformatics has proven invaluable for identifying genetic biomarkers and therapeutic targets; however, it is not without limitations that warrant critical consideration. A significant challenge lies in the dependency on predictive models and databases, which may yield inconsistent outcomes due to algorithmic variations and discrepancies in data quality. These tools frequently evaluate the functional consequences of genetic variants indirectly, necessitating experimental validation to establish their biological significance. Another constraint is the inability of bioinformatics approaches to fully encapsulate the complexity of gene-environment interactions. Factors such as lifestyle, microbiome diversity, and epigenetic influences on gene expression and therapeutic responses are often underrepresented in computational analyses. Additionally, the reliability of findings heavily depends on the accuracy and completeness of annotated genetic databases, which may differ across populations, potentially hindering the applicability of results to diverse demographic groups. Moreover, bioinformatics often concentrates on isolated genetic variants or pathways, which may oversimplify the intricate and multifactorial nature of diseases like IBD. The integration of combined genetic variations, epigenetic effects, and network-level interactions into computational models remains a considerable challenge. Further complicating the predictive utility of SNPs is the tendency to overlook critical factors, such as pharmacokinetics, pharmacodynamics, and individual variability in immune responses. Addressing these limitations necessitates complementary experimental investigations, large-scale population studies, and holistic methodologies that integrate bioinformatics predictions with empirical evidence to enhance the reliability and translational value of genetic findings in IBD research. Building on this study findings, future research should aim to validate the observed associations in larger, multi-ethnic cohorts to enhance the generalizability of the results. Functional studies are also needed to elucidate the biological mechanisms by which these polymorphisms influence on therapy response, including their impact on receptor signaling, cytokine production, and immune cell behavior. Integrating genomic data with transcriptomic and proteomic profiling could further refine predictive models for treatment outcomes. Additionally, exploring gene-environment interactions and the role of the gut microbiome may uncover complementary biomarkers that enhance therapeutic stratification. Ultimately, these efforts could pave the way for a personalized medicine framework in IBD management, optimizing drug selection and improving long-term patient outcomes. The present study provides a comprehensive elucidation of the genetic underpinnings of treatment response in patients with IBD. By unraveling the complexities of genetic variations and their influence on therapeutic efficacy, this research paves the way for the development of personalized treatment regimens that are tailored to the unique genetic profiles of individual patients. The application of such personalized approaches is anticipated to enhance clinical outcomes and significantly improve the quality of life for IBD patients. Specifically, our findings highlight the critical role of genetic polymorphisms, including the rs1799945 variant of the HFE gene and SNPs within the TNFRSF1B locus, in modulating patient responses to anti-TNF biologic therapies. These genetic markers serve as pivotal determinants in predicting treatment efficacy and optimizing therapeutic strategies. Unlike previous studies that focused primarily on European cohorts, this research expands the genetic landscape by exploring SNP-treatment associations in a Middle Eastern population. It is imperative to underscore that the results derived from this study are based on computational analyses, and therefore warrant validation through subsequent laboratory-based investigations. In conclusion, our study underscores the necessity for integrating genetic insights into clinical practice, with the ultimate objective of fostering precision medicine in the management of IBD. Future studies are essential to validate these computational findings and further explore the potential of genetic variations in enhancing treatment outcomes for IBD patients.