Generative Artificial Intelligence in Bioinformatics: A Systematic Review of Models, Applications, and Methodological Advances

TL;DR

This systematic review analyzes how generative AI, especially domain-specific language models and multimodal architectures, is transforming bioinformatics across genomics, proteomics, transcriptomics, and multi-omics integration. It finds that domain-tailored pretraining, tokenization, and fine-tuning outperform general-purpose LLMs on biological tasks, with multimodal fusion further boosting accuracy and generalization. The study highlights notable applications in de novo protein design, regulatory sequence interpretation, and interactive single-cell analysis, while also identifying limitations in data quality, scalability, interpretability, and grounding. It advocates a shift toward modular, efficient, knowledge-grounded GenAI systems that integrate verified bioinformatics tools to enable reliable, scalable discovery and translation to practice.

Abstract

Generative artificial intelligence (GenAI) has become a transformative approach in bioinformatics that often enables advancements in genomics, proteomics, transcriptomics, structural biology, and drug discovery. To systematically identify and evaluate these growing developments, this review proposed six research questions (RQs), according to the preferred reporting items for systematic reviews and meta-analysis methods. The objective is to evaluate impactful GenAI strategies in methodological advancement, predictive performance, and specialization, and to identify promising approaches for advanced modeling, data-intensive discovery, and integrative biological analysis. RQ1 highlights diverse applications across multiple bioinformatics subfields (sequence analysis, molecular design, and integrative data modeling), which demonstrate superior performance over traditional methods through pattern recognition and output generation. RQ2 reveals that adapted specialized model architectures outperformed general-purpose models, an advantage attributed to targeted pretraining and context-aware strategies. RQ3 identifies significant benefits in the bioinformatics domains, focusing on molecular analysis and data integration, which improves accuracy and reduces errors in complex analysis. RQ4 indicates improvements in structural modeling, functional prediction, and synthetic data generation, validated by established benchmarks. RQ5 suggests the main constraints, such as the lack of scalability and biases in data that impact generalizability, and proposes future directions focused on robust evaluation and biologically grounded modeling. RQ6 examines that molecular datasets (such as UniProtKB and ProteinNet12), cellular datasets (such as CELLxGENE and GTEx) and textual resources (such as PubMedQA and OMIM) broadly support the training and generalization of GenAI models.

Figures (12)

• Figure 1: Overview of the categorization framework of the review based on three research question dimensions: WHAT describes the bioinformatics domains and biological data modalities where GenAI is applied, WHY explains the goals for applying GenAI, and HOW outlines the AI approaches and tasks.
• Figure 2: A visual outline of the sections, indicating the flow from introduction and methodology through RQ answers, discussion, limitations, and future directions.
• Figure 3: Inclusion and exclusion process used for selecting articles based on predefined criteria such as publication year, domain relevance, and methodological quality.
• Figure 4: Year-wise distribution of selected articles across Q1 and Q2 journals between 2021 and 2025, highlighting the growing trend of GenAI research in bioinformatics.
• Figure 5: Geographic distribution of reviewed publications, illustrating the global research contributions from major countries and regions in the field of GenAI for bioinformatics.