Emerging machine learning (ML) technologies have the potential to significantly improve the research and treatment of rare diseases, which constitute a vast set of diseases that affect a small proportion of the total population. Artificial Intelligence (AI) algorithms can help to quickly identify patterns and associations that would be difficult or impossible for human analysts to detect. Predictive modeling techniques, such as deep learning, have been used to forecast the progression of rare diseases, enabling the development of more targeted treatments. Moreover, AI has also shown promise in the field of drug development for rare diseases with the identification of subpopulations of patients who may be most likely to respond to a particular drug. This review aims to highlight the achievements of AI algorithms in the study of rare diseases in the past decade and advise researchers on which methods have proven to be most effective. The review will focus on specific rare diseases, as defined by a prevalence rate that does not exceed 1–9/100,000 on Orphanet, and will examine which AI methods have been most successful in their study. We believe this review can guide clinicians and researchers in the successful application of ML in rare diseases.

Integrating Artificial Intelligence (AI) into Precision Medicine (PM) is redefining healthcare, enabling personalized treatments tailored to individual patients based on their genetic code, environment, and lifestyle. AI’s ability to analyze vast and complex datasets, including genomics and medical records, facilitates the identification of hidden patterns and correlations, which are critical for developing personalized treatment plans. Unsupervised Learning (UL) is particularly valuable in PM as it can analyze unstructured and unlabeled data to uncover novel disease subtypes, biomarkers, and patient stratifications. By revealing patterns that are not explicitly labeled, unsupervised algorithms enable the discovery of new insights into disease mechanisms and patient variability, advancing our understanding of individual responses to treatment. However, the integration of AI into PM presents some challenges, including concerns about data privacy and the rigorous validation of AI models in clinical practice. Despite these challenges, AI holds immense potential to revolutionize PM, offering a more personalized, efficient, and effective approach to healthcare. Collaboration among AI developers and clinicians is essential to fully realize this potential and ensure ethical and reliable implementation in medical practice. This review will explore the latest emerging UL technologies in the biomedical field with a particular focus on PM applications and their impact on human health and well-being.

Computational methods have transformed target and drug discovery, significantly accelerating the identification of biological targets and lead compounds. Despite its limitations, in silico molecular docking represents a foundational tool. Molecular Dynamics (MD) simulations, employing accurate force fields, provide near-realistic insights into a compound’s behavior within a biological target. However, docking and MD predictions may be unreliable without precise knowledge of the target binding site. Through MD simulations, we investigated 100 co-crystal structures of biological targets complexed with active compounds, identifying key structural and energy dynamic features that govern target–ligand interactions. Our analysis provides a detailed quantitative description of these parameters, offering critical validation for improving the predictive reliability of docking and MD simulations. This work provides a robust framework for refining early-stage drug discovery and target identification.

Background: Rare diseases represent a significant research challenge due to the limited availability of data, small patient cohorts, and heterogeneous phenotypes. Data augmentation and synthetic data generation are increasingly adopted to mitigate these limitations. Methods: This scoping review maps the application of data augmentation and synthetic data generation methods as strategies to address these limitations. A total of 118 studies published between 2018 and 2025 were identified through PubMed, Scopus, and Electronics Engineers (IEEE) Xplore. Results: Imaging data headed the field, followed by clinical and omics datasets. Classical augmentation, mainly geometric and photometric transformations, emerged as the most frequent approach, while deep generative models have rapidly expanded since 2021. Rule- and model-based methods were less common but demonstrated high interpretability in small datasets. Conclusions: Overall, these techniques enabled dataset expansion and improved model robustness. However, both approaches require rigorous validation to confirm biological plausibility. Together, these methods can transform data scarcity from a barrier into a driver of methodological innovation, enabling more inclusive rare disease research.

Alkaptonuria (AKU) is a rare metabolic disorder characterized by the accumulation of homogentisic acid (HGA), leading to progressive ochronosis and joint degeneration. While much is known about HGA’s role in tissue damage, the molecular mechanisms underlying acute inflammation in AKU remain poorly understood. Serum amyloid A (SAA) proteins are key mediators of the inflammatory response, yet their potential as biomarkers for inflammation in AKU has not been explored. This study investigated the role of the SAA1.1 allele as a biomarker for the severity of acute inflammation in AKU. Data from the ApreciseKUre Precision Medicine Ecosystem were analyzed to assess the relationship between SAA1 allelic variants and inflammatory markers. Molecular dynamics simulations compared the structural dynamics of SAA1.1 and SAA1.2 isoforms, with standard modeling and analysis pipelines employed. Using a clinomics approach, we showed that AKU patients expressing the SAA1.1 allele have significantly higher acute inflammation-related markers. Extensive molecular dynamics simulations revealed that the SAA1.1 isoform lent high structural instability of the C-terminal domain, accelerating the formation of amyloid fibrils and exacerbating the inflammatory condition. These findings would identify the SAA1.1 allele as a novel genetic biomarker for the progression of secondary amyloidosis in AKU and its severity. Furthermore, new molecular insights into the inflammatory mechanisms of AKU were provided, suggesting potential therapeutic approaches aimed at stabilizing SAA1.1 protein and preventing amyloid fibril formation, with significant implications in AKU and precision medicine strategies for SAA-related diseases.

Alkaptonuria (AKU) is a rare inherited metabolic disorder characterized by progressive musculoskeletal damage, chronic pain, and functional heterogeneity. To better quantify this variability, we introduced the concept of the functional age gap, defined as the difference between chronological age and a data-derived estimate of functional age. The study included 134 patients with AKU from the ApreciseKUre database. Functional age was calculated by mapping Health Assessment Questionnaire Disability Index (HAQ-DI) and Knee Injury and Osteoarthritis Outcome Score (KOOS) values to age-referenced normative data. Most patients (94.8%) showed a positive functional age gap, with a mean difference of 15 years, which indicates a functionally older profile than expected for their chronological age. A bagging ensemble of decision trees was then used to explore relationships between clinical variables and functional age gap severity. The model achieved moderate but stable classification performance across repeated stratified cross-validation (64%), consistent with an exploratory analysis in a small rare-disease cohort. SHapley Additive exPlanations analysis identified age, AKUSSI spinal pain, AKUSSI joint pain, Schober test, and hip and knee activity as the most influential predictors. These findings support the functional age gap as an interpretable, hypothesis-generating descriptive metric for functional assessment in AKU, while its predictive utility for individual patient stratification will require validation in larger and longitudinal cohorts.