IEEE/ACM Transactions on Computational Biology and Bioinformatics

Understanding which disease genes are altered by a drug can provide insight into the biology of effect, help us understand adverse drug effects, and suggest new drug uses. Here, we build on our model Draphnet in a new formulation with a similar goal. Draphnet was designed to explain drug therapeutic and side effects by learning a network connecting drugs to the disease genes they alter. Our new model, DraPhormer, has a similar goal but instead of relying on a linear model, learning of drug to gene connections uses a transformer model. DraPhormer integrates drug molecular data, disease genetics, and known drug effects on diseases, along with language models representing all of these entities. We show in simulations that DraPhormer can explain the genetic mechanisms of drug effects. Then, we present our design for incorporating drug and disease biology into the model. Finally, we benchmark the models ability to learn drug indications and side effects in real data.

Ovarian cancer is one of the gynecological cancer types, which, if metastasized and not detected early, can cause deaths among women. Therefore, there is a need to accurately predict drug responses to ovarian cancer. A gynecological pathologist inspects abnormality in tissues, followed by providing a report about patients; however, such a diagnostic process is (1) hard; (2) requires experience; and (3) time consuming. Moreover, existing tools are far from perfect. Hence, we present a computational pipeline to improve predicting drug response pertaining to ovarian cancer, derived as follows. First, we download digital pathology images pertaining to ovarian bevacizumab response from the cancer imaging archive repository. We employed histogram of oriented gradients to images, constructing feature vectors, provided to Fisher linear discriminant analysis to change the representation through dimensionality reduction. Then, we provide reduced-dimensionality data for regression analysis through support vector regression coupled with various kernels and calculating the area under the ROC curve (AUC). Experimental results against transformer-based models (ViT and Swin) and other deep learning (DL) models (VGG16, ResNet50, InceptionV3, MobileNetV2, and EfficientNetB6) demonstrate that our approach with radial kernel (named SVRD+R) yielded an AUC performance improvements of 17% against the best-performing transformer-based model (ViT) while obtaining an AUC performance improvements of 14.9% when compared against the best DL-based model (MobileNetV2). These results demonstrate the superiority and feasibility of our AI-based pipeline when tackling prediction problems pertaining to gynecologic cancer studies. MSC92B05; 68T09

Severe fetal growth restriction (sFGR) affects 5 to 10% of pregnancies worldwide and is a major contributor to perinatal morbidity and mortality, particularly in low- and middle-income countries (LMICs). Traditional 2D ultrasound detection methods suffer from operator dependency, gestational age uncertainty, and limited access to Doppler in many low-resource facilities. This study presents a deep learning framework for sFGR screening and triage using 2D fetal abdominal ultrasound images designed to operate independently of precise gestational dating. Growth restriction severity labels were derived by mapping abdominal circumference measurements to INTERGROWTH-21st term percentiles as a gestational-age-normalized proxy for fetal size restriction when case-level gestational age or birth-weight data are unavailable. A systematic literature review of 37 studies revealed gaps in severity stratification and generalizability. We implemented a DenseNet-121-based model with abdominal circumference measurement for severity-aware classification using a retrospective single-center dataset of 1588 annotated fetal abdominal images from 169 term pregnancies. Patient-wise 3-fold cross-validation and ensemble testing yielded 93.7% accuracy, a weighted F1-score of 0.76, and ROC AUC [≥] 0.98 per class on heldout data. The approach outperforms previously reported single-center methods on this dataset while explicitly targeting LMIC-specific constraints. It demonstrates potential as a gestational-age-independent first-line triage layer for equitable prenatal screening, subject to prospective multi-site validation.

Predicting protein-protein interactions is a fundamental problem in molecular biology. Experimental approaches for identifying protein-protein interactions are time-consuming and labor-intensive, motivating the development of efficient computational alternatives, including machine learning-based methods. However, conventional machine learning methods often rely on manually engineered features that require substantial domain expertise. In this study, we propose a two-stage framework to address these limitations. In the first stage, a one-dimensional convolutional neural network autoencoder is used to automatically learn latent representations from protein sequences. The quality of these features is evaluated through reconstruction error, reflecting how accurately the model reconstructs the original sequence. In the second stage, these learned features are combined with amino acid frequency-based features to form a hybrid feature set for predicting protein-protein interactions. A systematic comparison is performed between models trained on frequency features alone and those using a hybrid representation. The comparison showed that incorporating one-dimensional convolutional neural network-derived latent features improved the models performance of predicting protein-protein interactions. The dataset was split into training, validation, and test sets. Nested cross-validation was employed, with inner loops for hyperparameter tuning and outer loops for model selection. The random forest classifier achieved the best performance, with a mean receiver operating characteristic-area under curve of 0.91 and a test F1-score of 0.87. These results highlight the effectiveness of integrating deep feature learning with ensemble methods for predicting protein-protein interactions and build upon previous work focused on this fundamental problem. Author SummaryProtein-protein interactions are fundamental in all biological processes. However, predicting these interactions is a key problem in molecular biology. Computational approaches have been tested to address this problem. We applied a mix of machine learning and deep learning to gain insight into the qualities of proteins that engage in interaction. First, we trained a deep learning model, which automatically learned the primary sequence and characters related thereto, reducing bias in the actual prediction process. We combined these features, or latent representations, with amino acid frequency features of protein sequences, and called the two together "hybrid features." Then we performed a systematic comparison of amino acid frequency features-only with hybrid features, among four different machine learning classifiers. Our results suggest that the random forest classifier performed best among all four classifiers at predicting interactions between proteins. We propose that this approach could be used to improve efficiency in testing protein-protein interactions at the bench and may have applications to other biologically relevant molecular interactions.

Peripheral artery disease (PAD) is a complex vascular disorder characterized by heterogeneous molecular mechanisms and incomplete functional annotation, limiting systematic biomarker discovery. Network-based learning approaches provide a powerful framework for disease gene prioritization; however, most existing methods produce overconfident predictions without explicitly accounting for model uncertainty or structural novelty. Here, we present an uncertainty-aware framework for PAD biomarker discovery that integrates unsupervised graph representation learning, positive-unlabeled (PU) classification, ensemble prediction, and mechanistic explainability. Node embeddings were learned using multiple unsupervised graph neural network (GNN) objectives and combined with heterogeneous classifiers to generate ensemble-averaged probability estimates and epistemic uncertainty. By jointly modeling predictive confidence and embedding-space novelty, we stratified candidates into high-confidence rediscoveries and structurally novel hypotheses under explicit uncertainty control. Across eight embedding objectives and five classifiers, ensemble aggregation produced stable, well-calibrated predictions and enabled prioritization of 100 candidate PAD-associated proteins. Probability-heavy candidates clustered tightly with known PAD proteins and were enriched for established vascular and hemostatic pathways, including extracellular matrix organization, integrin signaling, coagulation, and fibrinolysis. In contrast, novelty-heavy candidates occupied distinct embedding-space regions and partitioned into multiple coherent clusters enriched for upstream regulatory and signaling processes, including G protein-coupled receptor, ephrin receptor, kinase-driven, and NF-{kappa}B-associated pathways. Five-fold cross-validated comparison with established PU learning baselines demonstrated consistent improvement across all evaluation metrics (AUC 0.916 {+/-} 0.019 vs. 0.821 {+/-} 0.030 for the best baseline), and external validity was confirmed by significant enrichment of top candidates for related cardiovascular disease annotations (5.7x above background). Together, these results demonstrate that integrating uncertainty, novelty, and explainability enables calibrated and biologically grounded biomarker prioritization, with broad applicability to PAD and other complex diseases. Author summaryPeripheral artery disease affects millions of people worldwide but remains underdiagnosed, partly because we lack reliable molecular markers to detect it early. In this study, we developed a computational framework that uses protein interaction network data to predict which proteins may be involved in PAD, even when we only know a small number of confirmed disease-associated proteins. Our approach combines graph neural network embeddings with a machine learning technique called positive-unlabeled learning, which is specifically designed for situations where you have confirmed positives but no confirmed negatives. We also quantify how confident the model is in each prediction and identify candidates that are genuinely novel compared to what is already known. Tested against established methods, our framework consistently found more known disease proteins in cross-validated evaluation. The candidates we identified map to biologically coherent pathways relevant to vascular disease, and our top predictions are enriched for proteins associated with related cardiovascular conditions, providing external validation. This work provides a principled and transparent approach to biomarker discovery that could be applied to other complex diseases with limited molecular annotations.

Whole-body SPECT bone scintigraphy reflects skeletal metabolic activity throughout the body and plays an indispensable role in the screening, treatment evaluation, and prognostic assessment of bone metastases in tumors. However, the automatic detection and segmentation of hypermetabolic bone lesions remain challenging due to low contrast, limited spatial resolution, and complex lesion distributions. In this study, we proposed Bone-Segnet, a dual-view guided automatic segmentation network for hypermetabolic bone lesions that integrated multi-scale feature modeling, global context modeling, and view-conditioned modulation. Pixel-level annotated anterior and posterior whole-body bone scintigraphy images were used for model training and prediction. The proposed network enhanced the recognition of low-contrast and small-scale lesions through small-lesion enhancement and multi-scale contextual modeling. A Transformer module was further introduced to strengthen global feature representation, while cross-view collaborative modeling was achieved by incorporating the complementary characteristics of anterior and posterior imaging. Experimental results demonstrated that the proposed method outperformed existing approaches across multiple evaluation metrics, with the Dice score improving from 0.7440 to 0.8750, indicating a substantial improvement in segmentation performance. Further quantitative analysis based on the segmentation results revealed significant differences among disease types in lesion count, pixel burden, and spatial distribution patterns, reflecting the heterogeneity of disease-related skeletal metabolic activity. Overall, the proposed method improved automatic lesion segmentation performance and enabled quantitative analysis of lesion burden and spatial distribution patterns, providing objective data support for the assessment of related diseases. Index Terms--Whole-body SPECT, bone lesion segmentation, dual-view modeling, quantitative analysis.

We developed a lightweight codon-based DNA Transformer equipped with multi-head self-attention and an adaptive classifier head, which achieves exon intron classification with high accuracy and also has moderate accuracy in CDS classification and splice site recognition. We named this model as ExIT (Exon-Intron Transformer). We have implemented codon tokenization for this model. This has been validated on the human genome with external validation from the chimpanzee genome. Further benchmarking has implied that our model is better than the existing models in the above tasks.

MotivationMetagenomic sequencing generates petabyte-scale sequence datasets that strain both deep learning and alignment based enzyme annotation tools. A lightweight rapid and accurate filter tool is needed to identify enzymatic sequences prior to resource-intensive functional prediction. ResultsWe present sxRaep (Rapid and Accurate Enzyme Predictor), a resource-efficient framework using lightweight physicochemical features for enzyme pre-screening. sxRaep achieves 6,604-fold speedup over Diamond (0.002 seconds per inference) with 62.1% memory reduction relative to Diamond (372 MB peak), while maintaining 99.4% accuracy and the highest recall in remote homology detection. This lightweight approach identifies enzymatic candidates missed by alignment-based methods without sacrificing accuracy. Availability and ImplementationsxRaep is available as a Python package at https://pypi.org/project/raep/, is maintained as an open-source software repository at https://github.com/labxscut/sxRaep, and can be deployed using the Docker image cirinmok/raep:python3.11 (https://hub.docker.com/r/cirinmok/raep/tags), which provides a reproducible Python 3.11 environment for enzyme prediction and model execution. Contactlcxia@scut.edu.cn

Pairwise alignment scores are used to classify pairs of sequences in many areas of bioinformatics, including homology search, predicting interactions, or read mapping. The relative scores of different pairs strongly depend on the choice of a substitution matrix and gap penalties, but the existing approaches for the estimation of these parameters do not directly optimize them for the task of classification. In this work, we present DiscrimAlign, a statistical model for discriminative learning of substitution matrices and gap penalties from a dataset of positive and negative pairs of unaligned biological sequences. The model links the alignment score of a sequence pair with the associated binary label through a logistic function and learns the parameters by likelihood maximization. We analyze theoretical properties of the model, derive and implement a learning procedure, study its performance in simulated experiments, and apply it to predict microRNA-target interactions. We show that sequence alignment with discriminative substitution matrices and gap penalties predicts the interactions comparably to state-of-the-art neural network classifiers while being more interpretable. An implementation of the model and reproducibility workflows are available at https://github.com/BioGeMT/DiscrimAlign.

Cellular reprogramming is a complex interplay between perturbations and regulatory elements, culminating in gene expression changes. Current computational approaches do not explicitly model these regulatory interactions. Here, we performed combinatorial reprogramming with cardiac transcription factors, followed by Multiome Perturb-Seq to measure perturbations, open chromatin, and gene expression in individual cells. We then developed PEPR-GNN (Perturbation-Enhancer-Promoter-RNA Graph Neural Network), a theoretical and computational framework to model regulome responses during complex genetic perturbations. By statistically associating gene regulatory relationships, PEPR-GNN organizes genes into regulomes with shared gene regulatory responses to reprogramming, including easy-to-reprogram cardiac genes, difficult-to-reprogram fibroblast genes, and context-specific genes where the impact of a reprogramming factor depends on the presence of others. Finally, we use PEPR-GNN for in silico modeling of how genetic modifications of enhancers can be used to tune gene responses to reprogramming. Overall, through the use of causal perturbation information and an enhancer-aware regulome model of gene regulation, PEPR-GNN can effectively model complex cellular responses to perturbation. HighlightsO_LIMultiome Perturb-Seq of GHMT reprogramming in MEFs with RNA/ATAC-Seq readout. C_LIO_LIPEPR-GNN: a computational framework to model perturbation-induced regulomes. C_LIO_LIPEPR-GNN aids the interpretation of regulomes by diverse reprogramming responses. C_LIO_LIPEPR-GNN enables in silico perturbation to tune gene responses to reprogramming. C_LI

Scientific models are widely used across the natural sciences as an interface between scientific theories and empirical data [1]. Such models play a key role, for example, in the study of human and animal learning, where they express algorithmic hypotheses and relate them to psychology and neuroscience data [2, 3]. These models are traditionally handcrafted by expert researchers based on existing theory or new insights. Such handcrafted models, however, are now known to fall short of capturing the full richness of behavior, even in their narrow domains [4-7]. An alternative data-driven approach has emerged, seeking to discover new insights by fitting and interpreting flexible models [8-11]. However, these tools require substantial human effort to derive insight from data, and it has been unclear how to discover new ideas from data efficiently. Here, we present DataDIVER, a general approach for automatically discovering computational models from data, and demonstrate that these models surface novel mechanistic insights into human and animal learning. Our approach delivers models that take the form of short computer programs, which are optimized both to fit data well and to be simple. These programs explicitly connect with existing theoretical frameworks and are readily understandable by human scientists. They can also be used to make novel predictions, some of which we show are borne out in re-analysis of existing data. General-purpose tools for surfacing new ideas from data, especially in combination with the large datasets that are increasingly available in many fields, stand to dramatically accelerate scientific discovery.

Counterfactual Explanation (CE) algorithms have been successfully applied to uncover the main factors driving computational diagnostic and prognostic predictions on tabular medical data. Recently, a new Network Medicine paradigm has been introduced for patient diagnosis and prognosis using Patient Similarity Networks (PSNs), i.e. graphs where patients are represented as nodes and their clinical and biomolecular similarities as edges. In this context, graph-based algorithms, including Graph Neural Networks (GNNs), can provide predictions using not only individual patient features but also their relations within a network of clinically and biomolecularly similar individuals. In this work, we propose the first CE algorithm tailored to explain diagnostic and prognostic predictions within PSNs. Alongside a contrastive GNN backbone, we introduce a versatile, model-agnostic counterfactual search method compatible with any underlying classifier. Preliminary results on synthetic data and on a cohort of patients affected by the Alzheimers disease show that our algorithm is competitive both with seminal tabular based CE algorithms and GNNExplainer, a well-established method for explaining graph-based classification tasks.

Drug-target affinity (DTA) prediction is a key task in drug discovery, enabling the estimation of the interaction strength between candidate compounds and biological targets. However, current models rely on connectivity-based molecular representations and do not explicitly account for the spatial organization, also known as stereochemistry. This limitation becomes evident when considering chirality, where a drug can exist as enantiomers, i.e., molecules that share the same atoms and bonds but differ in their three-dimensional arrangement. Despite their chemical similarity, they can interact differently with the same target, leading to variations in binding affinity and biological activity. In this paper, we propose a stereochemistry-aware DTA prediction framework that incorporates this information into molecular representations. Drug representations are learned from chemical structure using a directed-bond message passing graph neural network that captures enantiomers configurations, while protein targets are represented through sequence-based embeddings. Experiments on the Davis dataset demonstrate that our model can improve affinity prediction. Importantly, a case study on a manually curated dataset of enantiomers with different biological action shows that the model is able to distinguish the affinities in the two forms consistent with their experimentally observed biological activity. These findings support the relevance of stereochemistry-aware molecular representation for more accurate and chemically faithful DTA prediction.

Identifying causal relationships and distinguishing them from associations is a central scientific endeavor with many applications; knowing causal links between genes and diseases, for instance, can focus drug discovery on curing diseases beyond just symptom management. Despite several studies on automatically extracting relations between entities from large biomedical literature corpora like PubMed, only a few studies extract causal relations from abstracts and even fewer summarize corpus-level evidence for causal links. Recently, Large Language Models (LLMs) have been increasingly deployed to summarize biomedical information and extract relations; however, there is a distinct lack of explicit benchmarking comparing these generalized LLM-based methods against specialized, domain-aware frameworks for corpus-wide causal inference. In this work, we develop a method to infer Corpus-Wide Causal Score (CWCS) of a gene-disease (G-D) pair by integrating two pieces of evidence: (i) network-based causal signals in a prior gene regulatory network, quantified as a CWCS-Net score using an existing multilayer network centrality algorithm; and (ii) corpus-wide literature evidence, quantified as a CWCS-TD (TD for Truth Discovery) score using a newly-developed TD algorithm. Our CWCS-TD (scoring) algorithm jointly and iteratively estimates causal scores for multiple G-D pairs while modeling the reliability of PubMed abstracts co-mentioning them; and represents an advance in the field of TD algorithms due to its incorporation of bibliometric features of publications to address the challenge of sparsity of abstracts that assert a G-D causal relation. Using OMIM as an external expert-curated reference to evaluate classifications of G-D pairs as causal or not, our CWCS method achieved a causal class F1 score of 0.600 across ten diseases, outperforming both LLMs, GPT-4o and MMed-Llama 3 (this performance trend also persists when using area under the precision-recall curve as the evaluation metric). Both LLMs exhibit high recall accompanied by comparatively low precision, resulting in lower causal class F1 scores (0.505 for GPT-4o and 0.522 for MMed-Llama 3) due to large number of false positive predictions. Taken together, these evaluations and other ablation studies show the promise of our carefully designed algorithm in collating and integrating evidence of biomedical causal relations from both network- and literature-based sources, thereby supporting its broader applicability.

The rapid advancement of genome sequencing technologies has led to the accumulation of a vast number of protein sequences in public databases. However, a significant proportion of these proteins remain functionally uncharacterized. Concurrently, the expansion of protein sequence data has enabled the development of protein language models (pLMs). By distilling billions of years of evolutionary history into a latent representational space, these models have acquired an unprecedented capacity to predict both the tertiary structures and functions of proteins. In this study, we developed a deep learning-based method to predict whether two proteins catalyze the same enzymatic reaction. Our approach leverages latent representations generated by ESM Cambrian (ESM C), a state-of-the-art pLM, which are then processed through a neural network architecture integrating an attention mechanism. Our method outperformed existing approaches, including those based solely on full-length sequence similarity. Notably, it also surpassed our previous LightGBM-based model, which relied on structural similarity scores derived from AlphaFold-predicted models. Analysis of the attention weights reveals that our model autonomously highlights biologically significant sites, such as catalytic and binding residues. This demonstrates that integrating pLMs with attention mechanisms can enhance the accuracy and interpretability of protein function prediction while eliminating the need for manual feature engineering.

In cancer genomics, recurrent patterns of mutual exclusivity within a gene set can indicate shared biological context and involvement in tumorigenesis. However, existing methods are not designed to distinguish between mutual exclusivity arising from meaningful biological interactions from those influenced by heterogeneity between underlying patient subpopulations. In this work, we introduce MOSAIC, a novel statistical framework that models patient subgroup heterogeneity in mutual exclusivity analyses. In experiments with simulated data and real data from The Cancer Genome Atlas, we show that MOSAIC amplifies subgroup-specific mutual exclusivity signals, including between IDH1 and IDH2 in young low grade glioma patients, while reducing the effect of signals produced by underlying subgroup structures, such as distinct genomic lineages associated with histological subtypes of endometrial cancer. Finally, we demonstrate that MOSAIC is more powerful than existing p-value combination methods for patient subgroup stratification. MOSAIC is available as an open-source tool at https://github.com/reynalab/mosaic.

Aptamers are single stranded DNA or RNA molecules selected for their high affinity and specificity to bind target molecules, similar to antibodies. They are commonly selected through the SELEX process, which involves the iterative exposure of a random sequence library to a target and retaining the sequences showing good binding properties. To improve Lyme disease detection, we propose designing aptamers that specifically bind to the CspZ protein on the surface of Borrelia burgdorferi, the bacterium responsible for the disease. Starting with a SELEX process consisting of thirteen rounds, from which selected in vitro sequence candidates have emerged, we aim to propose a holistic process that selects in silico new sequence candidates that are further validated experimentally. Our approach relies on 1) using Machine Learning (ML) techniques, specifically a Restricted Boltzmann Machine (RBM), to digitally replicate the last round of the SELEX process, 2) integrating insights from text analysis methods, such as word2vec and n-grams, into the RBM model trained on the final-round SELEX dataset to represent and compare newly generated sequences with in vitro candidates, 3) selecting in silico sequences with strong potential to bind to CspZ protein, 4) experimentally validating the selected in silico sequences of step 3. Our holistic approach combines biological insights with statistical models to improve the efficiency and outcome of the SELEX process. We enhance the RBM model, designed to replicate the distribution of the final SELEX round, by integrating geometric representations of sequences, which is especially advantageous when dealing with limited datasets relative to the vast sequence space. In addition, it provides in silico sequence candidates with strong binding properties.

Accurate modeling of genotype-by-environment (GxE) interaction is critical for genomic prediction in plant breeding but remains challenging due to complex interaction structures. Conventional models often use the Hadamard product of genotype and environment covariance matrices to capture joint similarity, which may not fully represent GxE complexity. Here we propose a novel framework that derives covariance structures from the matrix multiplication of genotype and environment kernels, decomposing these into symmetric components incorporated as random effects in mixed models. Evaluated for 11 wheat and rice multi-environment datasets and across, this approach consistently outperformed the traditional Hadamard-based model, improving prediction accuracy by up to 13.2% in Pearsons correlation and enhancing top-selection accuracy. Combining both methods yielded the highest performance, indicating complementary information capture. This framework offers a flexible, interpretable, and computationally feasible extension for modeling GxE interaction, potentially enhancing genomic selection effectiveness under diverse environmental conditions.

We systematically evaluate and compare the performance of classical statistical methods (ARIMA), mechanistic compartmental models (SEIR), modern deep learning architectures (LSTM, DLinear, Autoformer), and an emerging time-series foundation model (TabPFN-TS) to forecasts the incidence of Influenza-Like Illness (ILI) across nine European countries. The models are benchmarked against a naive baseline and a multi-model ensemble (RespiCast) created by an initiative of the ECDC. In line with the operational practice of existing forecasting hubs, our entire evaluation is explicitly optimized for short-term horizons (1 to 4 weeks ahead). Interestingly, we found that the foundation model TabPFN-TS allows for great zero-shot inference capabilities. Without any task-specific retraining, it successfully overcomes extreme data scarcity to consistently outperform all other individual architectures, frequently rivalling or surpassing the RespiCast ensemble. Our results highlight how deep learning architectures are severely constrained by extreme data scarcity, typical in epidemic forecasting, requiring targeted endogenous data augmentation to reduce predictive errors. Within the deep learning class of models, we observe that simpler architectures (such as DLinear and LSTM) frequently exhibit greater robustness and outperform complex, attention-based models (such as Autoformer) when data is constrained. Finally, our results show how a weighted ensemble, constructed by fusing all the models, delivers highly robust forecasts in all regions considered. Overall, our findings showcase the transformative potential of zero-shot foundation models in epidemic forecasting and confirm the importance of multi-model ensembles.

Principal Component Analysis (PCA) is a fundamental tool in human genetics, widely used to study population structure. However, the rapid growth of modern genomic datasets, which often exceed main memory capacity, renders traditional PCA methods infeasible, motivating out-of-core approaches. Prior work on out-of-core genomic PCA has focused primarily on optimizing the inherently compute-intensive numerical core, largely overlooking the stages of data I/O and preprocessing, which emerge as significant performance bottlenecks at tera-scale. Furthermore, existing approaches remain limited to shared-memory single-node architectures, lacking support for distributed multi-node environments. To address these limitations, we introduce DistPCA, the first distributed out-of-core framework for tera-scale genomic PCA, implemented as a C++ library and scalable across both single- and multi-node systems. Built on top of Message Passage Interface (MPI), the proposed framework employs multi-level data parallelism across the entire PCA pipeline, combining multiprocessing, multithreading, SIMD vectorization, and compute-transfer overlap, while remaining compatible with block-based methods that rely on associative operations. Extensive evaluation on real and synthetic datasets demonstrates near-linear scalability, achieving speedups of up to 58.2x and over 98% reduction in wall-clock time, while maintaining parallel efficiency above 82% and preserving accuracy in the recovered principal components.