Biology subjects

Paliouras, G.

Publications and source records attributed to Paliouras, G..

5 recordsLinked to original sources

bioRxiv2026

Interpretable Forecasting of Kidney Cancer Progression via Generative AI and Symbolic Reasoning

Predicting cancer stage progression from omics data, and deriving molecular insight into the mechanisms driving it, remains a major challenge, owing in part to the lack of adequate longitudinal data and the interpretability limitations of current forecasting models. Large cancer datasets such as TCGA capture patient profiles cross-sectionally rather than longitudinally, complicating timely treatment decisions as tumors become more invasive. Deep neural networks typically used for forecasting, such as LSTMs, compound this problem by remaining largely opaque and offering clinicians no straightforward way to audit their predictions. Clear cell renal cell carcinoma (ccRCC) illustrates the clinical stakes of both challenges. Five-year survival falls from over 94% at stage I to 28% at stage IV, yet early-stage tumors are often managed under active surveillance, a strategy constrained by sparse molecular evidence of progression risk. Detecting progression in time, meanwhile, demands forecasts clinicians can interpret and trust, not black-box predictions. We address both challenges by combining generative and symbolic AI: a Variational Autoencoder trained on bulk RNA-Seq profiles of 530 TCGA ccRCC patients generates synthetic pseudo-time trajectories that overcome the absence of longitudinal data, while a symbolic rule-induction framework (ASAL) learns finite-state automata from these trajectories, encoding stage transition as human-readable Boolean conditions over gene expression, which a complex event forecasting system (Wayeb) converts into probabilistic forecasts of stage advancement. An independent XGBoost classifier trained on real patients (F1 score = 0.71-0.81) shows a gradual early-to-late probability shift along the synthetic trajectories, absent in non-progressing control trajectories. Pathway enrichment of those trajectories reveals stage-dependent changes in established kidney cancer-related processes, including the TCA cycle and DNA repair. Finally, our symbolic forecaster nearly matches an LSTM baseline (macro F1 = 0.928 vs. 0.964), while additionally offering an inspectable rule set and a probability distribution over transition timing rather than a single opaque score. This work shows that generative and symbolic AI, paired together, can turn cross-sectional cohorts into a transparent, forecast-oriented framework for modeling disease progression, demonstrated here in ccRCC.

bioinformatics↗

bioRxiv2025

CPP2Vec: a Representation Learning Approach for Cell-Penetrating Peptides Prediction

BackgroundCell-penetrating peptides (CPPs) facilitate the delivery of a variety of therapeutic molecules across the plasma membrane, from small chemical substances to nucleic acid-based macromolecules, such as antisense oligonucleotides (ASOs). Among neutral ASOs, peptide nucleic acids (PNAs) and phosphorodiamidate morpholino oligomers (PMOs) have been extensively studied as potential medical treatments for Duchenne Muscular Dystrophy (DMD), a severe genetic disease that causes muscle degeneration progressively. Over the last few decades, many in silico methods have emerged to detect novel CPPs, counterbalancing the cost of wet-lab experiments. ResultsIn this study, we propose CPP2Vec, a Word2Vec-based CPP prediction method, where the Word2Vec technique is used to represent amino acid sequences of peptides. We developed three task-specific supervised machine learning models for CPP-Classification, Uptake-Efficiency and PMO-Delivery. The first two models were designed to determine if an unseen peptide is a CPP and to pre-dict its uptake efficiency, respectively, while the PMO-Delivery model predicts if a peptide could enhance the cellular delivery of a PMO-complex compared to its naked version. Furthermore, we explored an alternative approach using pre-trained protein-based Large Language Models (LLMs) - T5, BERT, and ESM-2 - to generate the embeddings, resulting in three task-specific models, namely CPP2LLM. A comparison of CPP2Vec and CPP2LLM with state-of-the-art CPP prediction tools is included, proving their significant predictive performance. ConclusionIn this research, we present a Machine Learning (ML)-based tool that introduces the use of the Word2Vec technique in the field of CPPs pre-diction. Notably, it stands out for not requiring any manual a priori feature engineering and for its ability to generalize without any changes between studied tasks. CPP2Vec is available for use at: https://github.com/SSvolou/CPP2Vec.

bioinformatics↗

bioRxiv2024

Machine learning methods applied to classify complex diseases using genomic data

Complex diseases pose challenges in disease prediction due to their multifactorial and polygenic nature. In this work, we explored the prediction of two complex diseases, multiple sclerosis (MS) and Alzheimers disease (AD), using machine learning (ML) methods and genomic data from UK Biobank. Different ML methods were applied, including logistic regressions (LR), gradient boosting decision trees (GB), extremely randomized trees (ET), random forest (RF), feedforward networks (FFN), and convolutional neural networks (CNN). The primary goal of this research was to investigate the variability of ML models in classifying complex diseases based on genomic risk. LR was the most robust method across folds and diseases, whereas deep learning methods (FFN and CNN) exhibited high variability. When comparing the performance of polygenic risk scores (PRS) with ML methods, PRS consistently performed at an average level. However, PRS still offers several practical advantages over ML methods. Despite implementing feature selection techniques to exclude non-informative and correlated predictors, the performance of ML models did not improve significantly, underscoring the ability of ML methods to achieve optimal performance even in the presence of correlated features due to linkage disequilibrium. Upon applying explainability tools to extract information about the genomic features contributing most to the classification task, the results confirmed the polygenicity of MS. The prevalence of HLA gene annotations among the top genomic features on chromosome 6 aligns with their significance in the context of MS. Overall, the highest-prioritized genomic variants were identified as expression or splicing quantitative trait loci (eQTL or sQTL) located in non-coding regions within or near genes associated with the immune response and MS. In summary, this research offers deeper insights into how ML models discern genomic patterns related to complex diseases.

bioinformatics↗

bioRxiv2022

BioASQ-QA: A manually curated corpus for Biomedical Question Answering

The BioASQ question answering (QA) benchmark dataset contains questions in English, along with golden standard (reference) answers and related material. The dataset has been designed to reflect real information needs of biomedical experts and is therefore more realistic and challenging than most existing datasets. Furthermore, unlike most previous QA benchmarks that contain only exact answers, the BioASQ-QA dataset also includes ideal answers (in effect summaries), which are particularly useful for research on multi-document summarization. The dataset combines structured and unstructured data. The material linked with each question comprise documents and snippets, which are useful for Information Retrieval and Passage Retrieval experiments, as well as concepts that are useful in concept-to-text Natural Language Generation. Researchers working on paraphrasing and textual entailment can also measure the degree to which their methods improve the performance of biomedical QA systems. Last but not least, the dataset is continuously extended, as the BioASQ challenge is running and new data are generated.

bioinformatics↗

bioRxiv2022

CRISPRedict: The case for simple and interpretable efficiency prediction for CRISPR-Cas9 gene editing

The development of the CRISPR-Cas9 technology has provided a simple yet powerful system for targeted genome editing. Compared with previous gene-editing tools, the CRISPR-Cas9 system identifies target sites by the complementarity between the guide RNA (gRNA) and the DNA sequence, which is less expensive and time-consuming, as well as more precise and scalable. To effectively apply the CRISPR-Cas9 system, researchers need to identify target sites that can be cleaved efficiently and for which the candidate gRNAs have little or no cleavage at other genomic locations. For this reason, numerous computational approaches have been developed to predict cleavage efficiency and exclude undesirable targets. However, current design tools cannot robustly predict experimental success as prediction accuracy depends on the assumptions of the underlying model and how closely the experimental setup matches the training data. Moreover, the most successful tools implement complex machine learning and deep learning models, leading to predictions that are not easily interpretable. Here, we introduce CRISPRedict, a simple linear model that provides accurate and inter-pretable predictions for guide design. Comprehensive evaluation on twelve independent datasets demonstrated that CRISPRedict has an equivalent performance with the currently most accurate tools and outperforms the remaining ones. Moreover, it has the most robust performance for both U6 and T7 data, illustrating its applicability to tasks under different conditions. Therefore, our system can assist researchers in the gRNA design process by providing accurate and explainable predictions. These predictions can then be used to guide genome editing experiments and make plausible hypotheses for further investigation. The source code of CRISPRedict along with instructions for use is available at https://github.com/VKonstantakos/CRISPRedict.

bioinformatics↗