Biology subjects

Iversen, P.

Publications and source records attributed to Iversen, P..

4 recordsLinked to original sources

bioRxiv2026

Leveraging Uncertainty Estimates for Drug Response Prediction in Cancer Cell Lines

Machine learning models for drug response prediction in cancer cell lines carry the potential to advance precision oncology by tailoring treatments to the molecular tumor profile. Their application is challenged by variability in prediction quality and distribution shifts between training and application. Uncertainty estimation provides more information on the predictive distribution than point estimates, enabling comprehensive decision support and downstream analysis of predictions. Yet, the most effective uncertainty estimator is domain-specific. In this work, we benchmark uncertainty-aware models for drug response prediction. We focus on epistemic uncertainty via ensemble agreement, and aleatoric uncertainty via distributional modeling, or both. We find that ensemble-based estimates are more sensitive to distribution shift and can flag out-of-distribution examples. In contrast, distributional models yield stronger prediction error reductions among high-confidence subsets. Despite higher computational cost, the combination can provide both advantages: an ensemble of neural networks that estimate a Gaussian predictive distribution can reduce the mean squared error by 64 percent when restricting predictions to the 10 percent most confident drug-cell line pairs, and reliably indicates distribution shifts and platform differences. Beyond benchmarking, probabilistic predictions can identify drugs whose uncertainty bounds overlap with therapeutically relevant ranges. We also show that uncertainty estimates enable a new dimension of model interpretability: by attributing predicted uncertainty to input features, we identify genes that signal unpredictability of drug response rather than sensitivity or resistance. We further demonstrate uncertainty-guided selection of measurements for active learning. In summary, including uncertainty in drug response prediction supports better-informed model application. The code is available at https://github.com/PascalIversen/LUDRP.

bioinformatics↗

bioRxiv2025

From Hype to Health Check: Critical Evaluation of Drug Response Prediction Models with DrEval

MotivationLarge-scale drug sensitivity screens have enabled training drug response prediction models based on cancer cell line omics profiles to facilitate personalized medicine. While model performances reported in the literature appear promising, no successful translation to the clinic has been reported. ResultsWe identify six primary obstacles to this: non-reproducible models, data leakage leading to poor generalization, pseudoreplication, biased evaluation metrics, missing ablation studies, and inconsistent viability data reducing comparability across models. Together, these issues lead to overly optimistic performance estimates of state-of-the-art models and make it challenging to track progress in the field. To address this, we present DrEval, a pipeline for unbiased and biologically meaningful evaluation of cancer drug response models. It includes baseline and literature models with consistent hyperparameter tuning, statistically sound evaluations, and cross-study benchmarks. DrEval enables ablation studies and publication-ready visualizations. It allows researchers to focus on model development without implementing their own evaluation protocol. We find that deep learning models barely outperform a naive model predicting the mean drug and cell line effects, while no complex model significantly outperforms properly tuned tree-based ensemble baselines in relevant settings. We advocate making our pipeline a standard benchmark for cancer drug response prediction, ensuring a clinically relevant and robust assessment. Availability and implementationDrEval consists of a Python package, available on PyPI (drevalpy) and GitHub (github.com/daisybio/drevalpy), and an accompanying nf-core pipeline (github.com/nf-core/drugresponseeval). All data is available on Zenodo (DOI: 10.5281/zenodo.12633909), preprocessing scripts on github.com/daisybio/preprocess_drp_data.

bioinformatics↗

bioRxiv2024

Optimizing ODE-derived Synthetic Data for Transfer Learning in Dynamical Biological Systems

BackgroundAccurate prediction of the temporal dynamics of biological systems is crucial for informing timely and effective interventions, e.g., in ecological or epidemiological contexts, or for treatment adjustments in therapy. While machine learning has proven its capabilities in generalizing the underlying non-linear dynamics of such systems, unlocking its predictive power is often restrained by the limited availability of large, curated datasets. To supplement real-world data, informing machine learning by transfer learning with synthetic data derived from simulations using ordinary differential equations has emerged as a promising solution. However, the success of this approach highly depends on the designed characteristics of the synthetic data. ResultsWe suggest scrutinizing these characteristics, such as size, diversity, and noise, of ordinary differential equation-based synthetic time series datasets. Here, we demonstrate how to systematically evaluate the influence of such design choices on transfer learning performance. We conduct a proof-of-concept study on three simple, but widely used systems and four real-world datasets. We find a strong interdependency between synthetic dataset size and diversity effects. Good transfer learning settings heavily rely on real-world data characteristics as well as the datas coherence with the dynamics of the model underlying the synthetic data. We achieve a performance improvement of up to 92% in mean absolute error for simulation-based transfer learning compared to non-informed deep learning. ConclusionsOur work emphasizes the relevance of carefully selecting properties of synthetic data for leveraging the valuable domain knowledge contained in ordinary differential equation models for machine-learning based predictions. The code is available at https://github.com/DILiS-lab/opt-synthdata-4tl.

bioinformatics↗

bioRxiv2016

A genetic risk score to guide age-specific, personalized prostate cancer screening

BackgroundProstate-specific-antigen (PSA) screening resulted in reduced prostate cancer (PCa) mortality in a large clinical trial, but due to a high false-positive rate, among other concerns, many guidelines do not endorse universal screening and instead recommend an individualized decision based on each patients risk. Genetic risk may provide key information to guide the decisions of whether and at what age to screen an individual man for PCa.\n\nMethodsGenotype, PCa status, and age from 34,444 men of European ancestry from the PRACTICAL consortium database were analyzed to select single-nucleotide polymorphisms (SNPs) associated with prostate cancer diagnosis. These SNPs were then incorporated into a survival analysis to estimate their effects on age at PCa diagnosis. The resulting polygenic hazard score (PHS) is an assessment of individual genetic risk. The final model was validated in an independent dataset comprised of 6,417 men with screening PSA and genotype data. PHS was calculated for these men to test for prediction of PCa-free survival. PHS was also combined with age-specific PCa incidence data from the U.S. population to generate a PCa-Risk (PCaR) age that relates a given mans risk to that of the population average. PHS and PCaR age were evaluated for prediction of positive predictive value (PPV) of PSA screening.\n\nFindingsPHS calculated from 54 SNPs was very highly predictive of age at PCa diagnosis for men in the validation set (p =10-53). PPV of PSA screening varied from 0.18 to 0.52 for men with low and high genetic risk, respectively. PHS modulates PCa-free survival curves by an estimated 20 years between men in the 1st or 99th percentiles of genetic risk.\n\nInterpretationPolygenic hazard scores give personalized genetic risk estimates and can inform the decisions of whether and at what age to screen a man for PCa.\n\nFundingDepartment of Defense #W81XWH-13-1-0391

genetics↗