Keywords
Catechol-O-Methyltransferase
Neural networks, computer
Pain
Temporomandibular joint disorders
Metrica
How to Cite
Abstract
Background: This comparative analysis aims to deepen our understanding of catechol-O-methyltransferase (COMT) gene variations and their potential role in Temporomandibular Joint (TMJ) pain mechanisms. Methods: To analyze COMT protein sequences relevant to TMJ pain, UniProt IDs P21964 and Q8WZ04 were utilized, identifying proteins with 90% and 50% sequence similarity. These sequences were processed using the Deepbio tool, a deep-learning platform for biological sequence analysis. FASTA sequences were downloaded, validated, and divided into training and test datasets through Deepbio. The datasets were partitioned into 80 percent training and 20 percent test to optimize hyperparameters and evaluate performance. Three sequence prediction models—GAT (graph attention networks), transformer, and BiLSTM—were trained and tested to assess their predictive accuracy for TMJ painrelated COMT sequences, employing a structured approach to fine-tune parameters. Results: The sequence prediction models demonstrated promising results in identifying functional COMT sequences associated with TMJ pain.GAT has the highest accuracy (0.885), followed by BiLSTM (0.855) and Transformer (0.830). BiLSTM achieves the highest sensitivity (0.870), indicating better performance in identifying positive cases. These models exhibited strong sensitivity and specificity in identifying TMJ pain-associated sequences, indicating their potential utility in pinpointing genetic risk factors for TMJ pain. Conclusion: This study demonstrates that GAT and BiLSTM models, in particular, can effectively predict COMT sequence variants associated with TMJ pain, providing valuable insights into the genetic basis of TMD. As these computational approaches continue to evolve, they hold promise for improving diagnostic and therapeutic strategies for TMJ pain, underscoring the role of computational tools in molecular biology research.
References
1. Nascimento TD, Yang N, Salman D, Jassar H, Kaciroti N, Bellile E, et al. µ-Opioid activity in chronic TMD pain is associated with COMT polymorphism. J Dent Res. 2019 Nov;98(12):1324-31. doi: 10.1177/0022034519871938.
2. Schwahn C, Grabe HJ, Schwabedissen HM, Teumer A, Schmidt CO, Brinkman C, et al. The effect of catechol-O-methyltransferase polymorphisms on pain is modified by depressive symptoms. Eur J Pain. 2012 Jul;16(6):878-89. doi: 10.1002/j.1532-2149.2011.00067.x. Epub 2011 Dec 19.
3. Meloto CB, Segall SK, Smith S, Parisien M, Shabalina SA, Rizzatti-Barbosa CM, at al. COMT gene locus: new functional variants. Pain. 2015 Oct;156(10):2072-83. doi: 10.1097/j.pain.0000000000000273.
4. Nackley AG, Diatchenko L. Assessing potential functionality of catechol-O-methyltransferase (COMT) polymorphisms associated with pain sensitivity and temporomandibular joint disorders. Methods Mol Biol. 2010;617:375-93. doi: 10.1007/978-1-60327-323-7_28.
5. Diatchenko L, Slade GD, Nackley AG, Bhalang K, Sigurdsson A, Belfer I, et al. Genetic basis for individual variations in pain perception and the development of a chronic pain condition. Hum Mol Genet. 2005 Jan;14(1):135-43. doi: 10.1093/hmg/ddi013. Epub 2004 Nov 10.
6. Bonato LL, Quinelato V, Cordeiro PCF, Vieira AR, Granjeiro JM, Tesch R, et al. Polymorphisms in COMT, ADRB2 and HTR1A genes are associated with temporomandibular disorders in individuals with other arthralgias. Cranio. 2021 Jul;39(4):351-61. doi: 10.1080/08869634.2019.1632406. Epub 2019 Jul 2.
7. Slade GD, Ohrbach R, Greenspan JD, Fillingim RB, Bair E, Sanders AE, et al. Painful temporomandibular disorder: decade of discovery from OPPERA studies. J Dent Res. 2016 Sep;95(10):1084-92. doi: 10.1177/0022034516653743.
8. Slade GD, Sanders AE, Ohrbach R, Bair E, Maixner W, Greenspan JD, et al. COMT diplotype amplifies effect of stress on risk of temporomandibular pain. J Dent Res. 2015 Sep;94(9):1187-95. doi: 10.1177/0022034515595043.
9. Zhang X, Hartung JE, Bortsov AV, Kim S, O'Buckley SC, Kozlowski J, et al. Sustained stimulation of β2- and β3-adrenergic receptors leads to persistent functional pain and neuroinflammation. Brain Behav Immun. 2018 Oct;73:520-32. doi: 10.1016/j.bbi.2018.06.017.
10. Phero A, Ferrari LF, Taylor NE. A novel rat model of temporomandibular disorder with improved face and construct validities. Life Sci. 2021 Dec;286:120023. doi: 10.1016/j.lfs.2021.120023.
11. Slade GD, Fillingim RB, Ohrbach R, Hadgraft H, Willis J, Arbes SJ Jr, et al. COMT genotype and efficacy of propranolol for TMD pain: a randomized trial. J Dent Res. 2021 Feb;100(2):163-70. doi: 10.1177/0022034520962733. Epub 2020 Oct 8.
12. Meloto CB, Bortsov AV, Bair E, Helgeson E, Ostrom C, Smith SB, et al. Modification of COMT-dependent pain sensitivity by psychological stress and sex. Pain. 2016 Apr;157(4):858-67. doi: 10.1097/j.pain.0000000000000449.
13. Xuan P, Zhan L, Cui H, Zhang T, Nakaguchi T, Zhang W. Graph triple-attention network for disease-related LncRNA prediction. IEEE J Biomed Health Inform. 2022 Jun;26(6):2839-49. doi: 10.1109/JBHI.2021.3130110.
14. Lai PT, Lu Z. BERT-GT: cross-sentence n-ary relation extraction with BERT and Graph Transformer. Bioinformatics. 2021 Apr;36(24):5678-85. doi: 10.1093/bioinformatics/btaa1087.
15. Tejani AS. To BERT or not to BERT: advancing non-invasive prediction of tumor biomarkers using transformer-based natural language processing (NLP). Eur Radiol. 2023 Nov;33(11):8014-6. doi: 10.1007/s00330-023-10224-y.
16. Zhao X, Zhao X, Yin M. Heterogeneous graph attention network based on meta-paths for lncRNA-disease association prediction. Brief Bioinform. 2022 Jan;23(1):bbab407. doi: 10.1093/bib/bbab407.
17. Xiang Z, Gong W, Li Z, Yang X, Wang J, Wang H. Predicting protein-protein interactions via gated graph attention signed network. Biomolecules. 2021 May;11(6):799. doi: 10.3390/biom11060799.
18. Liu Q, Long C, Zhang J, Xu M, Tao D. Aspect-aware graph attention network for heterogeneous information networks. IEEE Trans Neural Netw Learn Syst. 2024 May;35(5):7259-66. doi: 10.1109/TNNLS.2022.3213799.
19. Gu W, Gao F, Lou X, Zhang J. Discovering latent node Information by graph attention network. Sci Rep. 2021 Mar;11(1):6967. doi: 10.1038/s41598-021-85826-x.
20. Ma Y, Guo Z, Xia B, Zhang Y, Liu X, Yu Y, et al. Identification of antimicrobial peptides from the human gut microbiome using deep learning. Nat Biotechnol. 2022 Jun;40(6):921-31. doi: 10.1038/s41587-022-01226-0.
21. Singh V, Shrivastava S, Singh SK, Kumar A, Saxena S. StaBle-ABPpred: a stacked ensemble predictor based on biLSTM and attention mechanism for accelerated discovery of antibacterial peptides. Brief Bioinform. 2022 Jan;23(1):bbab439. doi: 10.1093/bib/bbab439.
22. Sharma R, Shrivastava S, Singh SK, Kumar A, Saxena S, Singh RK. Deep-ABPpred: identifying antibacterial peptides in protein sequences using bidirectional LSTM with word2vec. Brief Bioinform. 2021 Sep;22(5):bbab065. doi: 10.1093/bib/bbab065.
23. Yadalam PK, Ardila CM. Enhanced hierarchical attention networks for predictive interactome analysis of LncRNA and CircRNA in oral herpes virus. J Oral Biol Craniofac Res. 2025 May-Jun;15(3):445-53. doi: 10.1016/j.jobcr.2025.02.012.
24. UniProt Consortium. UniProt: the Universal Protein Knowledgebase in 2023. Nucleic Acids Res. 2023 Jan;51(D1):D523-D531. doi: 10.1093/nar/gkac1052.
25. Wang R, Jiang Y, Jin J, Yin C, Yu H, Wang F, et al. DeepBIO: an automated and interpretable deep-learning platform for high-throughput biological sequence prediction, functional annotation and visualization analysis. Nucleic Acids Res. 2023 Apr;51(7):3017-29. doi: 10.1093/nar/gkad055.
26. Wei L, Ye X, Xue Y, Sakurai T, Wei L. ATSE: a peptide toxicity predictor by exploiting structural and evolutionary information based on graph neural network and attention mechanism. Brief Bioinform. 2021 Sep;22(5):bbab041. doi: 10.1093/bib/bbab041.
27. Abdin O, Nim S, Wen H, Kim PM. PepNN: a deep attention model for the identification of peptide binding sites. Commun Biol. 2022 May;5(1):503. doi: 10.1038/s42003-022-03445-2.
28. Sharma R, Shrivastava S, Singh SK, Kumar A, Saxena S, Singh RK. Deep-AFPpred: identifying novel antifungal peptides using pretrained embeddings from seq2vec with 1DCNN-BiLSTM. Brief Bioinform. 2022 Jan;23(1): bbab422. doi: 10.1093/bib/bbab422.
29. Cruz D, Monteiro F, Paço M, Vaz-Silva M, Lemos C, Alves-Ferreira M, et al. Genetic overlap between temporomandibular disorders and primary headaches: a systematic review. Jpn Dent Sci Rev. 2022 Nov;58:69-88. doi: 10.1016/j.jdsr.2022.02.002.
30. Poluha RL, Soares FFC, Furquim BD, Canales GT, Fiamengui LMSP, Bonjardim LR, et al. Painful temporomandibular joint clicking: genetic point of view. J Oral Facial Pain Headache. 2022 Summer;36(3-4):229-35. doi: 10.11607/ofph.3115.
31. Mladenovic I, Krunic J, Supic G, Kozomara R, Bokonjic D, Stojanovic N, et al. Pulp sensitivity: influence of sex, psychosocial variables, COMT gene, and chronic facial pain. J Endod. 2018 May;44(5):717-21.e1. doi: 10.1016/j.joen.2018.02.002.
32. Kambur O, Männistö PT. Catechol-O-methyltransferase and pain. Int Rev Neurobiol. 2010;95:227-79. doi: 10.1016/B978-0-12-381326-8.00010-7.
33. Li J, Sun C, Cai W, Li J, Rosen BP, Chen J. Insights into S-adenosyl-l-methionine (SAM)-dependent methyltransferase related diseases and genetic polymorphisms. Mutat Res Rev Mutat Res. 2021 Jul-Dec;788:108396. doi: 10.1016/j.mrrev.2021.108396.
34. Khawaja SN, Scrivani SJ. Trigeminal autonomic cephalalgia and facial pain: a review and case presentation. J Oral Facial Pain Headache. 2019 Winter;33(1):e1-e7. doi: 10.11607/ofph.2143.
This work is licensed under a Creative Commons Attribution 4.0 International License.
Copyright (c) 2026 Ramya Suresh, Pradeep Kumar Yadalam, Ramya Ramadoss, Carlos M Ardila