Publications associées à An explainability framework for deep learning on chemical reactions exemplified by enzyme-catalysed reaction classification

Publications associées (50)

Directly optimizing for synthesizability in generative molecular design using retrosynthesis models

Synthesizability in generative molecular design remains a pressing challenge. Existing methods to assess synthesizability include heuristics-based metrics or retrosynthesis models which predict a synthetic pathway. By contrast, an explicit approach anchors ...

ROYAL SOC CHEMISTRY2025

Machine learning-enabled estimation of cardiac output from peripheral waveforms is independent of blood pressure measurement location in an in silico population

Ramin Mohammadi, Nikolaos Stergiopoulos, Georgios Rovas, Lydia Aslanidou, Sokratis Anagnostopoulos

Monitoring of cardiac output (CO) is a mainstay of hemodynamic management in the acutely or critically ill patient. Invasive determination of CO using thermodilution, albeit regarded as the gold standard, is cumbersome and bears risks inherent to catheteri ...

2025

On the added value of sequential deep learning for the upscaling of evapotranspiration

Devis Tuia

Estimating ecosystem-atmosphere fluxes such as evapotranspiration (ET) in a robust manner and at a global scale remains a challenge. Methods based on machine learning (ML) have shown promising results in achieving such upscaling, providing a complementary ...

2025

Retrieval of Surface Solar Radiation through Implicit Albedo Recovery from Temporal Context

Michael Lehning, Devis Tuia, Yael Frischholz

To retrieve Surface Solar Radiation (SSR) from satellite images, a baseline reflectance of the observed ground in clear-sky conditions, also called background reflectance, is necessary to distinguish atmospheric absorption and scattering effects from surfa ...

2025

Optimal enzyme utilization suggests that concentrations and thermodynamics determine binding mechanisms and enzyme saturations

Vassily Hatzimanikatis, Daniel Robert Weilandt, Asli Sahin

One of the main challenges hampering the development of kinetic models is the lack of kinetic parameters for many enzymatic reactions. Here, the authors introduce a framework to explore the catalytically optimal operating conditions of any complex enzyme m ...

Nature Portfolio2023

Enhancing skin lesion classification: a CNN approach with human baseline comparison

Zaffar Ahmed Shaikh

This study presents an augmented hybrid approach for improving the diagnosis of malignant skin lesions by combining convolutional neural network (CNN) predictions with selective human interventions based on prediction confidence. The algorithm retains high ...

PeerJ2025

Expanding computational biochemistry towards complex and non-natural compounds

Anastasia Sveshnikova

In the beginning was the metabolism. The biochemical processes that make life possible transformed the soup of chemicals into the life on Earth we know today. Since then, living organisms have evolved, and life on Earth has become more complex. Living orga ...

EPFL2023

Integer Linear Programming for Unsupervised Training Set Selection in Molecular Machine Learning

Anne-Clémence Corminboeuf, Friedrich Eisenbrand, Rubén Laplaza Solanas, Puck Elisabeth van Gerwen, Ksenia Briling, Jan Weinreich, Matthieu Nicolas Haeberle

Integer linear programming (ILP) is an elegant approach to solve linear optimization problems, naturally described using integer decision variables. Within the context of physics-inspired machine learning (ML) applied to chemistry, we demonstrate the relev ...

IOP Publishing Ltd2025

Refining S-acylation: Structure, regulation, dynamics, and therapeutic implications

Françoise Gisou van der Goot Grunberg, Muhammad Umair Anwar

S-acylation is one of the most frequent posttranslational modifications that regulates diverse cellular processes. Anwar and van der Goot discuss the complexity and disease implications of this highly regulated reversible lipidation. ...

Rockefeller University Press2023

Combining physics-based and data-driven models: advancing the frontiers of research with scientific machine learning

Alfio Quarteroni

Scientific Machine Learning (SciML) is a recently emerged research field which combines physics-based and data-driven models for the numerical approximation of differential problems. Physics-based models rely on the physical understanding of the problem at ...

2025

Model-based scale-up of a continuously operated consolidated bioprocess based on a microbial consortium for the production of ethanol

David Beat Dempfle

This work demonstrated that consolidated bioprocessing is a promising concept for conversion of lignocellulose to ethanol at industrial scale. CBP offers great cost saving potential, is feasible to be operated continuously and may be scaled up due to exten ...

EPFL2022

A Study on Quantum Graph Neural Networks Applied to Molecular Physics

Su Yeon Chang

This paper introduces a novel architecture for Quantum Graph Neural Networks, which is significantly different from previous approaches found in the literature. The proposed approach produces similar outcomes with respect to previous models but with fewer ...

IOP Publishing Ltd2025

Dissecting the role of the asparagine synthesis in acute liver injury

Yu Sun

Amino acids are some of the main building stones of life used in a variety of biological processes, such as protein synthesis, energy metabolism or even as precursors for hormone synthesis. How-ever, some of them also participate and coordinate certain cel ...

EPFL2022

Predicting climate-change impacts on the global glacier-fed stream microbiome

Tom Ian Battin, Hannes Markus Peter, Tyler Joe Kohler, Grégoire Marie Octave Edouard Michoud, Massimo Bourquin, Matteo Tolosano, Vincent Henri De Staercke, Stylianos Fodelianakis, Andrew Lean Robison, Aileen Ute Geers

The shrinkage of glaciers and the vanishing of glacier-fed streams (GFSs) are emblematic of climate change. However, forecasts of how GFS microbiome structure and function will change under projected climate change scenarios are lacking. Combining 2,333 pr ...

NATURE PORTFOLIO2025

Investigation of Carboxylic Acid Isosteres and Prodrugs for Inhibition of the Human SIRT5 Lysine Deacylase Enzyme

Alexander Lund Nielsen

Sirtuin 5 (SIRT5) is a protein lysine deacylase enzyme that regulates diverse biology by hydrolyzing epsilon-N-carboxyacyllysine posttranslational modifications in the cell. Inhibition of SIRT5 has been linked to potential treatment of several cancers but ...

Wiley-VCH Verlag GmbH2022

Machine Learning in Decision-Making Systems: Fairness, Robustness, and Data Bias

Yves Rychener

Machine learning is increasingly integrated with decision making by powering automated systems and informing human decision makers in applications such as CV screening, portfolio allocation, loan risk assessment, and insurance approval. In this thesis, I e ...

EPFL2025

Characterization of the electron-accepting part of the organohalide respiration chain of Dehalobacter and Desulfitobacterium

Lorenzo Cimmino

Organohalide respiration (OHR) is a bacterial anaerobic process that use halogenated compounds, e.g. tetrachloroethene (PCE), as terminal electron acceptors. D. restrictus strain PER-K23, an obligate OHR bacterium (OHRB), and D. hafniense strain TCE1, a ba ...

EPFL2022

GLoFool: Global Enhancements and Local Perturbations to Craft Adversarial Images

Andrea Cavallaro

Adversarial examples crafted in black-box scenarios are affected by unrealistic colors or spatial artifacts. To prevent these shortcomings, we propose a novel strategy that generates adversarial images with low detectability and high transferability. The p ...

Springer Science and Business Media Deutschland GmbH2025

Modeling non-covalent interactions in condensed phase

Veronika Juraskova

The modeling of non-covalent interactions, solvation effects, and chemical reactions in complex molecular environment is a challenging task. Current state-of-the-art approaches often rely on static computations using implicit solvent models and harmonic ap ...

EPFL2022

SPOCK Tool for Constructing Empirical Volcano Diagrams from Catalytic Data

Anne-Clémence Corminboeuf, Rubén Laplaza Solanas, Romain Martin J Graux

Volcano plots, stemming from the Sabatier principle, visualize descriptor-performance relationships, allowing rational catalyst design. Manually drawn volcanoes originating from experimental studies are potentially prone to human bias as no guidelines or m ...

2025