A Systematic Review: Deep Learning for Analyzing Genomic Data to Discover Evolutionary Patterns

2025 · doi:10.22541/au.174584549.90714325/v1
preprint OA: closed
📄 Open PDF Full text JSON View at publisher
⚙ AI-generated summary by claude@2026-07, 2026-07-15 ⓘ

This systematic review found that convolutional neural networks are the most common deep learning method for genomic data analysis, with accuracy increasing to over 93% by 2025 due to architectural advances and hybrid models.

One-sentence paraphrase of the abstract; not a substitute for reading it. No clinical advice. How this works

⚙ AI-generated deep summary by claude@2026-07, 2026-07-15 · read from full text ⓘ

This preprint systematic review (Authorea, 2025) examined articles from 2020 to 2025 in which deep learning models were applied to genomic data to discover evolutionary patterns and to support related genomic analyses. Across the included studies, convolutional neural networks were reported as the most commonly used approach, with other architectures such as RNN/LSTM and GNN used in specific settings like time-series analysis, gene expression prediction, and molecular interaction discovery; reported model accuracy increased from about 88% in 2020 to over 93% in 2025, attributed to improved architectures, larger/higher-quality datasets, and hybrid models such as CNN+LSTM and GNN+CNN. The review emphasizes remaining challenges including multi-omics integration, explainable AI, fusion of genomic and image data, and transfer learning, and notes that the work is a preprint without peer review with potentially preliminary findings. Relevance to endometriosis: the paper does not explicitly discuss endometriosis or adenomyosis; it was included in the corpus via a keyword match in the upstream search index.

Read from the paper's body, not the abstract. Not a substitute for reading the paper. No clinical advice. How this works

Abstract

With the advancement of genetic sequencing technologies and the increase in the volume of biological data, deep learning (DL) has been adopted as one of the advanced data analysis methods in computational biology and genomic data analyzing. This review studied the articles published between 2020 and 2025 in which deep learning models were used to analyze genomic data. The results of this study show that convolutional neural networks (CNN) have been most widely used in genomic data analysis, while RNN, LSTM, and GNN methods have also been used in some applications such as time series data analysis, gene expression prediction, and molecular interaction discovery. The average accuracy of deep learning models in 2020 was about 88%, which increased to more than 93% in 2025. This improvement has been driven by advances in deep learning architectures, increased quality and volume of genomic data, and the use of hybrid models such as CNN+LSTM and GNN+CNN. In addition, most of the research has focused on cancer diagnosis, genetic diseases, and multi-omics data analysis. However, challenges remain, including multi-omics integration, explainable AI, fusion of genomic and image data, and transfer learning, which require further research. The results of this study indicate that deep learning in genomics data analysis can play an important role in discovering genetic patterns, improving the accuracy of disease diagnosis, and developing personalized therapeutic approaches. In the future, the development of multimodal deep learning models, interpretable methods (XAI), and the use of GNN for the analysis of gene networks and chromosomal interactions could lead to significant advances in this field.
Full text 2,753 characters · extracted from oa-doi-fallback · 2 sections · click to expand

Abstract

With the advancement of genetic sequencing technologies and the increase in the volume of biological data, deep learning (DL) has been adopted as one of the advanced data analysis methods in computational biology and genomic data analyzing. This review studied the articles published between 2020 and 2025 in which deep learning models were used to analyze genomic data. The results of this study show that convolutional neural networks (CNN) have been most widely used in genomic data analysis, while RNN, LSTM, and GNN methods have also been used in some applications such as time series data analysis, gene expression prediction, and molecular interaction discovery. The average accuracy of deep learning models in 2020 was about 88%, which increased to more than 93% in 2025. This improvement has been driven by advances in deep learning architectures, increased quality and volume of genomic data, and the use of hybrid models such as CNN+LSTM and GNN+CNN. In addition, most of the research has focused on cancer diagnosis, genetic diseases, and multi-omics data analysis. However, challenges remain, including multi-omics integration, explainable AI, fusion of genomic and image data, and transfer learning, which require further research. The results of this study indicate that deep learning in genomics data analysis can play an important role in discovering genetic patterns, improving the accuracy of disease diagnosis, and developing personalized therapeutic approaches. In the future, the development of multimodal deep learning models, interpretable methods (XAI), and the use of GNN for the analysis of gene networks and chromosomal interactions could lead to significant advances in this field. Supplementary Material File (final paper in english (2).docx) - Download - 864.49 KB Information & Authors Information Version history Copyright This work is licensed under a Non Exclusive No Reuse License.

Keywords

Authors Metrics & Citations Metrics Article Usage 314views 139downloads Citations Download citation Raha Hassanpour. A Systematic Review: Deep Learning for Analyzing Genomic Data to Discover Evolutionary Patterns. Authorea. 28 April 2025. DOI: https://doi.org/10.22541/au.174584549.90714325/v1 DOI: https://doi.org/10.22541/au.174584549.90714325/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu.

Text is read by the "Ask this paper" AI Q&A widget below. Extraction quality varies by source — PMC NXML preserves structure cleanly, OA-HTML may include some navigation residue, and OA-PDF can have broken hyphenation. The publisher copy (via DOI) is the canonical version.

My notes (saved in your browser only)

⚙ Ask this paper AI returns verbatim quotes from the full text · source: oa-doi-fallback ⓘ

Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works

Citation neighborhood (no data yet)

We don't have any in-corpus citations linked to this paper yet. This is a recent paper (2025) — citers typically take a year or two to land, and the OpenAlex reference graph may still be filling in.

Source provenance

europepmc
last seen: 2026-05-20T01:45:00.602351+00:00
unpaywall
last seen: 2026-06-02T02:00:03.124865+00:00