About

Raquel Dias is an Assistant Professor at the Department of Microbiology & Cell Science at the University of Florida. Her lab develops and applies cutting-edge Artificial Intelligence (AI) methods to examine important research questions across multiple fields of biological sciences and biomedical research.

Research topics

• Artificial Intelligence
• Computer Science
• Bioinformatics
• Statistics
• Medicine
• Mathematics
• Biology
• Genetics
• Computational biology
• Cognitive psychology
• Engineering
• Data science
• Psychology

Selected publications

• SpheronizaTor: Spherical Voxelization for Interpretable Protein Microenvironment Modeling

  Computational and Structural Biotechnology Journal · 2026-03-01

  articleOpen accessSenior author

  Artificial intelligence (AI) has expanded the reach of structural biology by enabling models to extract biochemical and geometric features directly from 3-dimensional (3D) protein structures. Yet, the effectiveness of these models depends critically on how protein environments are encoded. Most existing volumetric representations rely on Cartesian voxel grids derived from smoothed atomic densities, an approach that offers broad applicability but struggles to reconcile rotational invariance, residue-level specificity, and explicit biochemical detail. We present SpheronizaTor, a residue-centered voxelization framework for protein structures that builds local spherical voxel maps centered on each residue. Each spherical map encodes atom types, covalent bonding information, and whether atoms belong to the central residue or neighboring residues. By producing one voxel representation per residue, SpheronizaTor emphasizes the structural and functional granularity through which proteins organize catalysis, recognition, and stability. The combination of spherical alignment with chemically explicit feature channels enables richer interpretability and enhances compatibility with 3D convolutional and hybrid neural architectures. Designed specifically for proteins and engineered for extensibility, SpheronizaTor provides a voxelization strategy that is both chemically realistic and computationally efficient. The residue-centric approach bridges the gap between global volumetric encoders and graph-based models, offering a versatile foundation for downstream tasks such as mutation effect prediction, binding site analysis, and structural comparison across protein families.

  PublisherDOI

• The inflammatory path toward type 1 diabetes begins during pregnancy

  Nature Communications · 2026-01-07 · 2 citations

  articleOpen access

  Type 1 diabetes (T1D) is increasing globally, yet the earliest biological determinants remain poorly defined, particularly in general population studies. We studied the Swedish population-based ABIS birth cohort (n = 16,683) to identify early-life risk factors. Olink proteomic analysis (n = 286 controls, n = 146 cases) of inflammatory signals at birth shows differential abundance years before diagnosis (mean age 12.6 years), with proteins enriched for neutrophil migration, cytotoxicity, extracellular matrix remodeling, and immune regulation. Several markers remain significant in spite of prenatal and perinatal factors including family history of diabetes, and are associated with differences in compounds like stearic acid, lysine, glutamine, and persistent, environmental toxicants perfluorodecylethanoic acid and perfluorooctane sulfonate (PFOS). Using machine learning, we identify a protein subset that predicts T1D with high accuracy (AUC = 0.89 ± 0.02), independently of HLA genetic risk. These findings suggest that innate and tissue-remodeling pathways are perturbed at birth, possibly reflecting early β-cell vulnerability. Identifying these disruptions at birth with a non-invasive method opens a window for prevention, protecting β-cells before the inflammatory attack on islets begins.

  PublisherDOI

• A Phenome-Wide Comorbidity Atlas of Age-Related Hearing Loss, Speech-in-Noise Deficits, and Tinnitus: Distinguishing Causal Signals from Correlation

  Journal of the Association for Research in Otolaryngology · 2025-09-29 · 2 citations

  articleOpen access
  PublisherOA PDFDOI

• AI-Driven Personalization in Mobile Health Applications: An Elderly-Focused Approach to Health Monitoring and Prediction

  Advances in intelligent systems and computing · 2025-01-01

  book-chapter1st authorCorresponding
  PublisherDOI

• EquilibraTor streamlines molecular dynamics simulations in a single execution

  Computational and Structural Biotechnology Journal · 2025-01-01 · 2 citations

  articleOpen accessSenior author

  enzyme-substrate complex, a key biosynthetic step in natural product pathways with potential relevance for drug discovery. EquilibraTor offers a robust and user-friendly solution for integrating MD simulations into structural biology or drug discovery pipelines without extensive computational expertise.

  PublisherDOI

• Artificial intelligence‐powered plant phenomics: Progress, challenges, and opportunities

  The Plant Phenome Journal · 2025-12-29 · 1 citations

  articleOpen access

  Abstract Artificial intelligence (AI), a key driver of the Fourth Industrial Revolution, is being rapidly integrated into plant phenomics to automate sensing, accelerate data analysis, and support decision‐making in phenomic prediction and genomic selection. This perspective paper synthesizes current advances, identifies major barriers, and proposes future directions to realize the transformative potential of AI‐enabled plant phenomics. We first provide an overview of AI technologies with the potential to address key challenges in phenomics, from data collection to phenotypic trait extraction and environmental sensing. We then present three case studies focusing on specialty crops (blueberry [ Vaccinium corymbosum L.] mechanical harvestability traits, strawberry [ Fragaria × ananassa (Duchesne ex Weston)] production, and citrus [ Citrus L.] disease) to illustrate practical applications of AI‐driven phenomics. Moreover, we highlight future perspectives and opportunities for further research and innovation. These include large foundation models, real‐time inference on edge devices, explainable AI, generative AI and digital twins, AI‐enhanced multi‐omics, agentic AI, and knowledge‐guided and data‐driven hybrid approaches. Finally, we discuss key challenges and limitations of applying AI to plant phenomics, including data curation, model generalization and bias, and ethical considerations related to equitable access to AI tools.

  PublisherOA PDFDOI

• STICI: Split-Transformer with integrated convolutions for genotype imputation

  Nature Communications · 2025-01-31 · 8 citations

  articleOpen access

  Despite advances in sequencing technologies, genome-scale datasets often contain missing bases and genomic segments, hindering downstream analyses. Genotype imputation addresses this issue and has been a cornerstone pre-processing step in genetic and genomic studies. Although various methods have been widely adopted for genotype imputation, it remains challenging to impute certain genomic regions and large structural variants. Here, we present a transformer-based framework, named STICI, for accurate genotype imputation. STICI models automatically learn genome-wide patterns of linkage disequilibrium, evidenced by much higher imputation accuracy in regions with highly linked variants. Our imputation results on the human 1000 Genomes Project and non-human genomes show that STICI can achieve high imputation accuracy comparable to the state-of-the-art genotype imputation methods, with the additional capability to impute multi-allelic variants and various types of genetic variants. STICI can be trained for any collection of genomes automatically using self-supervision. Moreover, STICI shows excellent performance without needing any special presuppositions about the underlying patterns in collections of non-human genomes, pointing to adaptability and applications of STICI to impute missing genotypes in any species. Genome-scale genotyping datasets often contain missing data that negatively affects downstream analysis. Here, authors propose a genotype imputation method based on a Transformer framework that excels at imputing missing genotypes of genetic variants in various scenarios.

  PublisherOA PDFDOI

• Limitations of current machine learning models in predicting enzymatic functions for uncharacterized proteins

  G3 Genes Genomes Genetics · 2025-07-24 · 4 citations

  articleOpen access

  Thirty to seventy percent of proteins in any given genome have no assigned function and have been labeled as the protein "unknome." This large knowledge shortfall is one of the final frontiers of biology. Machine learning (ML) approaches are enticing, with early successes demonstrating the ability to propagate functional knowledge from experimentally characterized proteins. An open question is the ability of ML approaches to predict enzymatic functions unseen in the training sets. By integrating literature and a combination of bioinformatic approaches, we evaluated individually Enzyme Commission number predictions for over 450 Escherichia coli unknowns made using state-of-the-art ML approaches. We found that current ML methods not only mostly fail to make novel predictions but also make basic logic errors in their predictions that human annotators avoid by leveraging the available knowledge base. This underscores the need to include assessments of prediction uncertainty in model output and to test for "hallucinations" (logic failures) as a part of model evaluation. Explainable artificial intelligence analysis can be used to identify indicators of prediction errors, potentially identifying the most relevant data to include in the next generation of computational models.

  PublisherOA PDFDOI

• Challenges and Opportunities for New Frontiers and Technologies to Guarantee Food Production

  Sustainability · 2025-04-23 · 8 citations

  articleOpen accessCorresponding

  The global food production sector is under immense pressure due to rapid population growth and climate change, demanding innovative solutions for food security and sustainability. This review explores innovative advancements in agriculture and food technology, from urban farming (e.g., vertical farming, aquaponics, and hydroponics) to regenerative agriculture and agroforestry practices that enhance soil health and biodiversity. We also examine food production in extreme environments, including desert agriculture and space agriculture, as well as advances in biotechnology, synthetic biology, and nanotechnology, that enable improved crop yields and nutrition. The transformative role of AI in precision farming, predictive analytics, and water management is highlighted, as well as the importance of bioproducts and eco-friendly innovations. Finally, we discuss the vital role of policy, regulation, and community-driven approaches in shaping a resilient global food system. Through the integration of technology with sustainable practices, this review aims to inspire research into solutions that ensure future food security while preserving our planet.

  PublisherOA PDFDOI

• Limitations of Current Machine-Learning Models in Predicting Enzymatic Functions for Uncharacterized Proteins

  bioRxiv (Cold Spring Harbor Laboratory) · 2024-07-03 · 2 citations

  preprintOpen access

  Abstract Thirty to seventy percent of proteins in any given genome have no assigned function and have been labeled as the protein “unknome”. This large knowledge shortfall is one of the final frontiers of biology. Machine-Learning (ML) approaches are enticing, with early successes demonstrating the ability to propagate functional knowledge from experimentally characterized proteins. An open question is the ability of machine-learning approaches to predict enzymatic functions unseen in the training sets. Using a set of Escherichia coli unknowns, we evaluated the current state-of-the-art machine-learning approaches and found that these methods currently lack the ability to integrate scientific reasoning into their prediction algorithms. While human annotators can leverage the plethora of genomic data in making plausible predictions into the unknown, current ML methods not only fail to make novel predictions but also make basic logic errors in their predictions. This underscores the need to include assessments of prediction uncertainty in model output and to test for ‘hallucinations’ (logic failures) as a part of model evaluation. Explainable AI (XAI) analysis can be used to identify indicators of prediction errors, potentially identifying the most relevant data to include in the next generation of computational models. Article Summary Many proteins in any genome, ranging from 30% to 70% of the genome, lack an assigned function. This knowledge gap limits the full use of the vast available genomic data. Machine learning has shown promise in transferring functional knowledge within isofunctional families, but it largely fails to predict novel functions not seen in its training data. Understanding these failures can guide the development of better machine-learning methods to help experts make accurate functional predictions for uncharacterized proteins.

  PublisherOA PDFDOI

Frequent coauthors

• Ali Torkamani

  Scripps Health

  38 shared

• Shang‐Fu Chen

  National Taiwan University

  27 shared

• Nathan E. Wineinger

  Scripps Research Institute

  18 shared

• D. Gareth Evans

  The Christie NHS Foundation Trust

  17 shared

• Kaiyu Chen

  15 shared

• Jong Hun Lee

  13 shared

• Daria Prilutsky

  Takeda (United States)

  13 shared

• Sándor Szalma

  Takeda (United States)

  13 shared

Awards & honors

• UF/IFAS Archer Early Career Seed Grant (2025)