{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T07:16:58Z","timestamp":1788419818787,"version":"build-2803163510"},"reference-count":52,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T00:00:00Z","timestamp":1788393600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Computational classifiers can be benchmarked against public encoded splice-junction datasets, but the performance estimate will depend on the input representation, preprocessing pipeline, model set-up and evaluation design.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This study investigates the performance of 15 different Convolutional, Hybrid and Deep Learning architectures on a public encoded splice-junction dataset for 3-class classification. The results of the primary models were archived from one train\/validation\/held-out-test workflow and are presented as illustrative effects and not as statistically significant evidence of superiority.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The highest held-out test accuracy (96.87%) was obtained with the Custom CNN from among the configurations that were saved. The MLP Mixer was the most highly accurate model during training (99.90%), but the model did not perform as well on the validation set (93.33%) or held-out test set (93.73%), suggesting some in-sample fitting. Model-capacity information, parameter-to-sample ratios, learning curves, class-wise metrics, and a targeted class-weighting comparison are reported. Class-weighting was performed during the training phase only, but class-weighting did not yield highest performance for the tested Custom CNN configuration. of the chosen Custom CNN configuration for the observed dataset.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>The conclusions are limited to the classification results that could be computed for the encoded benchmark dataset and the reported evaluation procedure. The results do not demonstrate a statistically significant superiority of the models in the sense of external generalization, clinical diagnostic validity, patient-based mutation detection, clinical utility or readiness for deployment. Leakage-free repeated evaluation, external datasets, stored sample predictions, statistically comparison of models\u2019 outputs, and clinically validated patient-level data would be needed for more general conclusions.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/frai.2026.1886470","type":"journal-article","created":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T06:46:30Z","timestamp":1788417990000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["A comparative evaluation of deep learning models for the classification of encoded splice-junction sequences"],"prefix":"10.3389","volume":"9","author":[{"given":"Yogesh","family":"Kumar","sequence":"first","affiliation":[{"name":"Department of Computer Science and Engineering, School of Technology, Pandit Deendayal Energy University","place":["Gandhinagar, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Inderpreet","family":"Kaur","sequence":"additional","affiliation":[{"name":"Department of Computer Applications, Chandigarh Group of Colleges","place":["Mohali, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nandini","family":"Modi","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, School of Technology, Pandit Deendayal Energy University","place":["Gandhinagar, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Priya","family":"Bhardwaj","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, School of Computing, DIT University","place":["Dehradun, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jaeyoung","family":"Choi","sequence":"additional","affiliation":[{"name":"School of Computing, Gachon University","place":["Seongnam-si, Republic of Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Muhammad Fazal","family":"Ijaz","sequence":"additional","affiliation":[{"name":"School of Technology, Faculty of Business and Hospitality, Torrens University Australia","place":["Melbourne, VIC, Australia"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,3]]},"reference":[{"key":"ref1","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1016\/B978-0-443-27574-6.00005-9","volume-title":"Deep Learning in Genetics and Genomics","author":"Aarif","year":"2025"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"150377","DOI":"10.1109\/ACCESS.2020.3016651","article-title":"FCN-based DenseNet framework for automated detection and classification of skin lesions in dermoscopy images","volume":"8","author":"Adegun","year":"2020","journal-title":"IEEE Access"},{"key":"ref3","volume-title":"Towards the Next Generation of Clinical Decision Support: Overcoming the Integration Challenges of Genomic Data and Electronic Health Records","author":"Al Kawam","year":"2018"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"102","DOI":"10.5755\/j01.itc.50.1.25349","article-title":"An Efficient Technique for Disease Prediction by using Enhanced Machine Learning Algorithms for Categorical Medical Dataset","volume":"50","author":"Anusuya","year":"2021","journal-title":"Information Technology and Control"},{"key":"ref5","first-page":"4407","article-title":"Basic gene discretization-model using correlation clustering for distributed DNA databases","volume":"11","author":"Arputharaj","year":"2020","journal-title":"J. Adv. Res."},{"key":"ref6","doi-asserted-by":"publisher","first-page":"197","DOI":"10.1504\/IJBIDM.2021.112991","article-title":"CARs-RP: lasso-based class association rules pruning","volume":"18","author":"Azmi","year":"2021","journal-title":"Int. J. Business Intelligence Data Mining."},{"key":"ref7","doi-asserted-by":"publisher","first-page":"30505","DOI":"10.1007\/s11042-021-10944-7","article-title":"Recognition of splice-Junction genetic Sequences using Random forest and Bayesian Optimization","volume":"80","author":"Baareh","year":"2021","journal-title":"Multimed. Tools Appl."},{"key":"ref8","volume-title":"Integrating Bioinformatics and Clinical Insights: Towards a Comprehensive Framework for Precision Medicine","author":"Bianchi","year":"2024"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"1580","DOI":"10.3390\/antibiotics12111580","article-title":"Current uses and future perspectives of genomic technologies in clinical microbiology","volume":"12","author":"Bianconi","year":"2023","journal-title":"Antibiotics"},{"key":"ref10","first-page":"012004","volume-title":"Journal of Physics: Conference Series","author":"Borgalli","year":"2022"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"100426","DOI":"10.1016\/j.dajour.2024.100426","article-title":"A clinical named entity recognition model using pretrained word embedding and deep neural networks","volume":"10","author":"Dash","year":"2024","journal-title":"Decis. Anal. J."},{"key":"ref13","volume-title":"Deep Splicer: Predicting splice sites in the human Genome using Convolutional Blocks","author":"Fernandez-Castillo","year":"2022"},{"key":"ref14","volume-title":"IEEE 36th Conference on Machine Learning Applications","author":"Garouani","year":"2024"},{"key":"ref44","first-page":"329","article-title":"DNA Splice Junction Prediction Using Hybrid Approach of LSTM-GRU Model. In International Conference on Artificial Intelligence, Machine Learning and Intelligent Systems","author":"Gautam","year":"2025"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbaf705","article-title":"Detection of alternative splicing: deep sequencing or deep learning?","volume":"27","author":"Hackl","year":"2026","journal-title":"Briefings Bioinformatics"},{"key":"ref16","first-page":"1016867","article-title":"UniSplicer: a deep learning framework for accurate splice-site prediction and splice-altering mutation detection across diverse taxa. Plant","volume":"7","author":"Hong","year":"2026","journal-title":"Communications"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"5109","DOI":"10.1007\/s10462-022-10138-z","article-title":"Design possibilities and challenges of DNN models: a review on the perspective of end devices","volume":"55","author":"Hussain","year":"2022","journal-title":"Artif. Intell. Rev."},{"key":"ref18","volume-title":"DNA Sequencing using brain-Inspired Hyperdimensional Computing","author":"Imani","year":"2019"},{"key":"ref19","author":"Imani","year":"2018"},{"key":"ref48","doi-asserted-by":"crossref","first-page":"535","DOI":"10.1016\/j.cell.2018.12.015","article-title":"Predicting splicing from primary sequence with deep learning","volume":"176","author":"Jaganathan","year":"2019","journal-title":"Cell"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.37933\/nipes\/7.4.2025.SI268","article-title":"Gene fusion detection in cancer: integrating sequencing technologies and machine learning approaches","volume":"7","author":"Jonathan","year":"2025","journal-title":"NIPES J. Sci. Technol. Res."},{"key":"ref33","doi-asserted-by":"crossref","first-page":"156","DOI":"10.1007\/s11274-026-04891-4","article-title":"Gene regulatory and biomolecular networks and their multifaceted biotechnological applications","volume":"42","author":"Kar","year":"2026","journal-title":"World Journal of Microbiology and Biotechnology"},{"key":"ref21","first-page":"523","volume-title":"Computational Methods in Science and Technology","author":"Kaur","year":"2025"},{"key":"ref22","volume-title":"Developing a Heuristic Algorithm for Classification of Problems with Binary Attributes","author":"Konovalenko","year":"2019"},{"key":"ref23","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1007\/978-3-030-42227-1_5","volume-title":"Computational Intelligence: a Methodological Introduction","author":"Kruse","year":"2022"},{"key":"ref24","first-page":"1026","volume-title":"Seminars in Thrombosis and Hemostasis","author":"Kumar","year":"2024"},{"key":"ref25","volume-title":"Leveraging Functional Genomic Data: Modeling Transcriptional Dynamics and Interpreting Disease Genomes","author":"Lee","year":"2019"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"949","DOI":"10.1038\/s41588-024-02053-6","article-title":"Predicting RNA-seq coverage from DNA sequence as a unifying model of gene regulation","volume":"57","author":"Linder","year":"2025","journal-title":"Nat. Genet."},{"key":"ref27","article-title":"Enhanced structured lasso pruning with class-wise information","author":"Liu","year":"2025","journal-title":"arXiv preprint arXiv:2502.09125"},{"key":"ref28","volume-title":"2D Neural Network on 1D data. International Conference on Artificial Intelligence","author":"Loka","year":"2019"},{"key":"ref9001","doi-asserted-by":"crossref","first-page":"596","DOI":"10.3390\/bioengineering12060596","article-title":"Variable Selection for Multivariate Failure Time Data via Regularized Sparse-Input Neural Network","volume":"12","author":"Luo","year":"2025","journal-title":"Bioengineering"},{"key":"ref29","volume-title":"KINS: Knowledge Injection via Network Structuring","author":"Magnini","year":"2022"},{"key":"ref11","doi-asserted-by":"crossref","first-page":"61","DOI":"10.21037\/jmai-24-146","article-title":"Multimodal deep learning breast cancer prognosis models: narrative review on multimodal architectures and concatenation approaches","volume":"8","author":"Maigari","year":"2025","journal-title":"Journal of Medical Artificial Intelligence."},{"key":"ref30","first-page":"23886","article-title":"SRATTA: sample re-ATTribution attack of secure aggregation in federated learning","volume":"202","author":"Marchand","year":"2023","journal-title":"Proceed. Machine Learn. Res."},{"key":"ref31","article-title":"Learning decision trees with gradient descent","author":"Marton","year":"2023","journal-title":"arXiv preprint arXiv:2301.04567"},{"key":"ref37","author":"Nakagiri","year":"2025"},{"key":"ref32","volume-title":"Minimum Complexity echo state Networks for Genome and Sequence Analysis","author":"Neighbor","year":"2020"},{"key":"ref34","doi-asserted-by":"publisher","first-page":"997","DOI":"10.3390\/biology12070997","article-title":"Next-generation sequencing technology: current trends and advancements","volume":"12","author":"Satam","year":"2023","journal-title":"Biology"},{"key":"ref35","volume-title":"DNA Sequence Classification: An Advanced Machine Learning Framework for Accurate splice Junction Detection","author":"Sharif","year":"2025"},{"key":"ref36","first-page":"1","volume-title":"2024 11th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions)(ICRITO)","author":"Singh","year":"2024"},{"key":"ref38","doi-asserted-by":"publisher","first-page":"1583","DOI":"10.1007\/s10618-019-00635-1","article-title":"Efficient Mixture Model for Clustering of Sparse High-Dimensional Binary data","volume":"33","author":"\u015amieja","year":"2019","journal-title":"Data Mining Knowledge Discovery"},{"key":"ref39","doi-asserted-by":"publisher","first-page":"1242","DOI":"10.1109\/TNNLS.2019.2919608","article-title":"Completely automated CNN architecture design based on blocks","volume":"31","author":"Sun","year":"2019","journal-title":"IEEE Trans. Neural Networks Learn. Syst."},{"key":"ref40","doi-asserted-by":"crossref","DOI":"10.21203\/rs.3.rs-6414470\/v1","volume-title":"A Transformer-based Ensemble Approach for DNA splice Junction Classification","author":"Swaraja","year":"2025"},{"key":"ref41","doi-asserted-by":"publisher","first-page":"27","DOI":"10.17933\/jppi.v9i1.249","article-title":"Comparation Analysis of Ensemble Technique with Boosting (XGBoost) and Bagging (Random Forest) for Classifying splice Junction DNA Sequence Category","volume":"9","author":"Syahrani","year":"2019","journal-title":"Jurnal Penelitian Pos dan Informatika"},{"key":"ref42","first-page":"24261","article-title":"Mlp-mixer: an all-mlp architecture for vision","volume":"34","author":"Tolstikhin","year":"2021","journal-title":"Adv. Neural Inf. Proces. Syst."},{"key":"ref43","doi-asserted-by":"publisher","first-page":"6101","DOI":"10.3390\/ijms24076101","article-title":"Emerging personalized opportunities for enhancing translational readthrough in rare genetic diseases and beyond","volume":"24","author":"Wagner","year":"2023","journal-title":"Int. J. Mol. Sci."},{"key":"ref45","doi-asserted-by":"publisher","first-page":"3741","DOI":"10.1093\/bioinformatics\/btac359","article-title":"NanoSplicer: accurate identification of splice junctions using Oxford Nanopore sequencing","volume":"38","author":"You","year":"2022","journal-title":"Bioinformatics"},{"key":"ref46","doi-asserted-by":"publisher","first-page":"10221","DOI":"10.1038\/s41598-023-34795-4","article-title":"An automated framework for evaluation of deep learning models for splice site predictions","volume":"13","author":"Zabardast","year":"2023","journal-title":"Sci. Rep."},{"key":"ref47","doi-asserted-by":"crossref","DOI":"10.21203\/rs.3.rs-2256837\/v3","volume-title":"Single Molecule long-read eccDNA Sequencing Reveals Unambiguous and Selective Transposon Activation in Arabidopsis in Response to Stress, cellular state, and Epigenetic Processes","author":"Zaidi","year":"2023"},{"key":"ref51","doi-asserted-by":"publisher","first-page":"133","DOI":"10.1016\/j.isprsjprs.2017.07.014","article-title":"A hybrid MLP-CNN classifier for very fine resolution remotely sensed image classification","volume":"140","author":"Zhang","year":"2018","journal-title":"ISPRS J. Photogramm. Remote Sens."},{"key":"ref49","doi-asserted-by":"publisher","first-page":"e2057","DOI":"10.7717\/peerj-cs.2057","article-title":"Prediction model of stock return on investment based on hybrid DNN and TabNet model","volume":"10","author":"Zhang","year":"2024","journal-title":"PeerJ Computer Science"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.3390\/fractalfract7080588","article-title":"New Classifier Ensemble and Fuzzy Community Detection Methods using POP Choquet-like Integrals","volume":"7","author":"Zhang","year":"2023","journal-title":"Fractal Fractional"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1886470\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T06:46:37Z","timestamp":1788417997000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1886470\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,3]]},"references-count":52,"alternative-id":["10.3389\/frai.2026.1886470"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1886470","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,3]]},"article-number":"1886470"}}