2025 Hyper Recent •CC0 1.0 Universal

This work is dedicated to the public domain. No rights reserved.

Access Preprint From Server
May 9th, 2025
Version: 1
Sanofi Aventis DE
systems biology
biorxiv

Machine Learning Meets Pharmacokinetics: A Comparative Analysis of Predictive Models for Plasma Concentration-Time Profiles

Jost, F.Open in Google Scholar•Giegerich, C.Open in Google Scholar•Grebner, C.Open in Google Scholar•Matter, H.Open in Google Scholar•Cordes, H.Open in Google Scholar

Predicting pharmacokinetic (PK) profiles from molecular structures constitutes a significant advancement in pharmaceutical research with substantial implications for expediting the entire drug discovery process. Our investigation presents a comprehensive comparative analysis of five distinct methodological frameworks for predicting rat plasma concentration-time profiles: four approaches that integrate mechanistic modeling with computational techniques, and one approach employing pure machine learning (ML) architecture without physiological constraints. As reference (1), we used a method that predicts non-compartmental analysis (NCA) parameters to predict plasma PK profiles with a one-compartmental PK model (NCA-ML). Alternative approaches include: (2) a physiologically based PK (PBPK) model utilizing ML-predicted in vitro characteristics; (3) CMT-ML, where neural networks predict compartmental PK model parameters; (4) CMT-PINN, employing physics-informed neural networks; and (5) PURE-ML, using decision trees to predict concentration values at specific timepoints based on predicted volume of distribution and clearance. Assessment using multiple performance metrics demonstrated that the CMT-PINN approach achieved superior predictivity for PK profiles. Models trained directly on concentration-time, rather than derived PK parameters, delivered markedly improved predictive performance, particularly when working with limited training datasets. Our findings confirm the viability of predicting PK behavior from molecular structures before synthesis, representing a significant advancement for subsequent design. Implementation of these computational approaches enables informed compound selection early in the discovery pipeline, concentrating valuable resources for advanced preclinical investigations on only the most promising candidates, with anticipated beneficial effects on overall development timelines.

Similar Papers

biorxiv
Fri May 09 2025
Regulation of the branched-chain amino acid pathway and its crosstalk to carbon metabolism in yeast
In Saccharomyces cerevisiae, the TORC1 pathway regulates the transition from rapid fermentative growth to respiration during the diauxic shift, by tightly coordinating energy- and biomass-producing pathways. Although leucine and other components of the branched-chain amino acid (BCAA) pathway are known TORC1 regulators, how the BCAA pathway is controlled across this transition and influences the c...
Escalera-Fanjul, X.
•
Bolaji, O.
•
Drotleff, B.
•
DeLuna, A.
...•
Klipp, E.
biorxiv
Thu May 08 2025
Regulation of Transcriptional Bursting and Spatial Patterning in Early Drosophila Embryo Development
Nascent RNA synthesis often occurs in periods of high transcriptional activity, interspersed with basal or no activity periods. This phenomenon, known as transcriptional bursting, drives high intercellular variability in gene expression levels. How do key patterning genes in early Drosophila melanogaster embryos overcome this variability to establish precise spatial patterns for tissue development...
Nieto, C.
•
Vahdat, Z.
•
Lim, B.
•
Singh, A.
biorxiv
Thu May 08 2025
Virtual Colon: Spatiotemporal modelling of metabolic interactions in a computational colonic environment
Host-microbial metabolic interactions have been recognised as an essential factor in host health and disease. Genome-scale metabolic modelling approaches have made important contributions to our understanding of the interactions in such communities. One particular such modelling approach is BacArena in which metabolic models grow, reproduce, and interact as independent agents in a spatiotemporal m...
Marinos, G.
•
Zimmermann, J.
•
Taubenheim, J.
•
Kaleta, C.
biorxiv
Thu May 08 2025
A multilayer network approach elucidates time- and tissue-specific developmental and aging processes
Understanding the dynamic interplay of proteins across different life stages and tissues is essential for deciphering the molecular mechanisms underpinning development, aging, and disease. Here, we present a comprehensive network-based framework that constructs and integrates 119 time- and tissue-specific protein-protein interaction (PPI) networks derived from transcriptomic data, offering insight...
Lombardo, S. D.
•
Rendeiro, A. F.
•
Menche, J.
biorxiv
Thu May 08 2025
An integrative framework linking molecular signatures and locomotory phenotypes in space-induced sarcopenia
Age-related skeletal muscle deterioration, referred to as sarcopenia, poses significant risks to astronaut health and mission success during spaceflight, yet its multisystem drivers remain poorly understood. While terrestrial sarcopenia manifests gradually through aging, spaceflight induces analogous musculoskeletal decline within weeks, providing an accelerated model to study conserved atrophy me...
Ball, B. K.
•
Khan, H. F.
biorxiv
Wed May 07 2025
Identification and Overexpression of Endogenous Transcription Factors to Enhance Lipid Accumulation in the Commercially Relevant Species Chlamydomonas pacifica
Sustainable low-carbon energy solutions are critical to mitigating global carbon emissions. Algae-based platforms offer potential by converting carbon dioxide into valuable products while aiding carbon sequestration. However, scaling algae cultivation faces challenges like contamination in outdoor systems. Previously, our lab evolved Chlamydomonas pacifica, an extremophile green alga, which tolera...
Gupta, A.
•
Oliver, A.
•
Molino, J. V. D.
•
Wnuk-Fink, K. M.
...•
Mayfield, S.
biorxiv
Wed May 07 2025
Context-dependent spatial multicellular network motifs for single-cell spatial biology
The clinical state of diseased tissue is caused by complex intercellular processes that go beyond pairwise cell-cell interactions and are difficult to infer due to the combinatorial explosion of such high-dimensionality. We present context-dependent identification of spatial motifs (CISM), a two-step method to identify local cell structures associated with a disease state in single cell spatial da...
Zamir, A.
•
Amitay, Y.
•
Tamir, Y.
•
Keren, L.
•
Zaritsky, A.
biorxiv
Wed May 07 2025
The zoo of the gene networks capable of pattern formation by extracellular signaling
A fundamental question of developmental biology is pattern formation, or how cells with specific gene expression end up in specific locations in the body to form tissues, organs and, overall, functional anatomy. Pattern formation involves communication through extracellular signals and complex intracellular gene networks integrating these signals to determine cell responses (e.g., further signalin...
Anhon, K. M.
•
Ciudad, I. S.
biorxiv
Wed May 07 2025
EMMAi: fast enzyme-allocation constraints in GEMs for improved biomass prediction across carbon sources
Genome-scale metabolic models (GEMs) predict emergent phenotypes by modeling the metabolic networks encoded in genomes. While GEMs have significantly advanced systems biology, metabolic engineering, biomedicine, and environmental science, they require extensive time and resources for manual curation, which can limit their utility in rapidly evolving research landscapes. Recent findings suggest tha...
Molina Ortiz, J. P.
•
Benson, D.
•
Watts, J.
•
Velasque, M.
...•
Morgan, M.