Nature Computational Science

📢Generative AI code assistants show promise for scientific software development, but user overreliance risks undetected errors. In this Comment, Gabrielle O'Brien highlights key vulnerabilities and proposes actions for users.
www.nature.com/articles/s43...
🔓https://rdcu.be/eyky6
#compsky

📢In this month's editorial, we provide our recommendations for effectively writing computational science manuscripts.
www.nature.com/articles/s43...

White and grey topography

🚨Our July issue is now live and includes research on adverse drug reactions, multi-fidelity Bayesian optimization, rare event sampling, and much more!
t.co/myy3LVRjGO

Guidelines for multi-fidelity Bayesian optimization of molecules and materials - Nature Computational Science A recent study provides intuition and guidelines for deciding whether to incorporate cheaper, lower-fidelity experiments into a closed-loop search for molecules and materials with desired properties.

The corresponding News & Views from ‬ @corymsimon.bsky.social ‪and Qia Ke is also available!
‬ www.nature.com/articles/s43...
🔓https://rdcu.be/exyUz

Best practices for multi-fidelity Bayesian optimization in materials and molecular research - Nature Computational Science Multi-fidelity Bayesian optimization methods are studied on molecular and material discovery tasks, and guidelines are provided to recommend cheaper and informative low-fidelity sources when using thi...

Out now! @pschwllr.bsky.social and colleagues provide guidelines and recommendations for when to use multi-fidelity Bayesian optimization over their single-fidelity counterparts
www.nature.com/articles/s43...
🔓https://rdcu.be/exyUh
#chemsky

Learning committor-consistent collective variables - Nature Computational Science An artificial neural network-based strategy is developed to learn committor-consistent transition pathways, providing insight into rare events in biomolecular systems.

The corresponding News & Views from @gervasiolab.bsky.social is out now!
www.nature.com/articles/s43...
🔓https://rdcu.be/exl41

Privacy-preserving multicenter differential protein abundance analysis with FedProt - Nature Computational Science In this Resource, the authors present FedProt, a tool that enables privacy-preserving, federated differential protein abundance analysis across multiple institutions. Its results match the results of ...

Researchers from @cosybio-uhh.bsky.social present FedProt which enables privacy-preserving, federated differential protein abundance analysis across multiple institutions. #proteomics
www.nature.com/articles/s43...

Advancing neural decoding with deep learning - Nature Computational Science A recent study introduces a neural code conversion method that aligns brain activity across individuals without shared stimuli, using deep neural network-derived features to match stimulus content.

A News & Views is also available for this piece!
www.nature.com/articles/s43...
🔓https://rdcu.be/evPY6

Inter-individual and inter-site neural code conversion without shared stimuli - Nature Computational Science A neural code conversion method is introduced using deep neural network representations to align brain data across individuals without shared stimuli. The approach enables accurate inter-individual br...

Out now! @ykamit.bsky.social and colleagues present a neural code conversion method to align brain data across individuals without shared stimuli. The approach enables accurate inter-individual brain decoding and visual image reconstruction across sites. #compneurosky
www.nature.com/articles/s43...

Investigating the volume and diversity of data needed for generalizable antibody–antigen ΔΔG prediction - Nature Computational Science Predicting the effects of mutations on antibody–antigen binding is a key challenge in therapeutic antibody development. Orders of magnitude more data will be needed to unlock accurate, robust predicti...

Out now! @alissahummer.com @opig.stats.ox.ac.uk
and colleagues present Graphinity, a method to predict change in antibody-antigen binding affinity (∆∆G). Also featuring synthetic datasets of ~1 million FoldX-generated and >20,000 Rosetta Flex ddG-generated ∆∆G values!
www.nature.com/articles/s43...

Iterative variational learning of committor-consistent transition pathways using artificial neural networks - Nature Computational Science A neural network approach grounded in transition-path theory is shown to uncover committor-consistent transition pathways, resolving competing mechanisms across dynamical regimes and advancing the mod...

Out now! Chrisophe Chipot and colleagues develop a neural-network approach for committor-consistent enhanced-sampling simulations of rare events.
www.nature.com/articles/s43...
🔓https://rdcu.be/evcjW

Predicting adverse drug reactions for combination pharmacotherapy with cross-scale associative learning via attention modules - Nature Computational Science An associative-learning-enhanced model, called OrganADR, is developed to predict adverse drug reactions at the organ level. OrganADR could be applied in emerging combination pharmacotherapies and impr...

Out now! Xiaoqiong Li and colleagues develop an approach, OrganADR, to predict adverse drug reactions at the organ level for emerging combination pharmacotherapy
www.nature.com/articles/s43...
🔓https://rdcu.be/evcgl

Quantifying batch effects for individual genes in single-cell data - Nature Computational Science Group technical effects, a metric to quantify gene-level batch effects in single-cell data, is proposed. The analysis based on group technical effects reveals the unbalanced batch effects across genes...

Out now! Shuilin Jin and colleagues report a metric to quantify gene-level batch effects in single-cell data. 🖥️ 🧬
www.nature.com/articles/s43...
🔓 rdcu.be/etNNO

RNA–ligand interaction scoring via data perturbation and augmentation modeling - Nature Computational Science RNAsmol is a sequence-based deep learning framework for predicting RNA–small molecule interactions, integrating perturbation and augmentation to achieve robust predictions under limited data and adapt...

Out now! Zhi John Lu and colleagues present RNAsmol, a sequence-based framework for predicting RNA-small molecule interactions #RNASky
www.nature.com/articles/s43...
🔓 rdcu.be/etNJk

Molecule with blue pink and white lights, pink grid, blue network

🚨Our June issue is now live and includes research on cell-cell communication, a study on gender and racial diversity in science, and much more!
www.nature.com/natcomputsci...

@gpggrp.bsky.social @cmuscience.bsky.social @cmuengineering.bsky.social

Rethinking chemical research in the age of large language models - Nature Computational Science This Perspective highlights the potential integrations of large language models (LLMs) in chemical research and provides guidance on the effective use of LLMs as research partners, noting the ethical ...

A Perspective from @gabegomes.bsky.social and colleagues covers the opportunities for LLMs to advance chemical research and the challenges that must be overcome to effectively use LLMs as scientific partners.

www.nature.com/articles/s43...

🔓https://rdcu.be/etcqs

Cost-effective instruction learning for pathology vision and language analysis - Nature Computational Science Training foundation models often requires a costly budget and excessive computational resources. In this study, a low-cost instruction learning framework is proposed that could enable the rapid adopti...

📢Shaoting Zhang and colleagues propose CLOVER, a cost-effective instruction learning framework for conversational pathology. #bioimaging 🩺🖥️ www.nature.com/articles/s43...

🔓https://rdcu.be/erQld

A formally exact method for high-throughput absolute binding-free-energy calculations - Nature Computational Science A thermodynamic cycle is proposed that limits protein–ligand relative motion and boosts the efficiency of absolute binding-free-energy calculations four- to eightfold, without compromising theoretical...

📢Out now! Haohao Fu and colleagues introduce a high-throughput, formally exact method for absolute binding free energy calculations. #chemsky www.nature.com/articles/s43...

🔓https://rdcu.be/erBRm

Memory kernel minimization-based neural networks for discovering slow collective variables of biomolecular dynamics - Nature Computational Science This study presents MEMnets, an approach that integrates statistical mechanics with deep learning to identify the slowest collective variables for biomolecular dynamics. MEMnets effectively captures m...

📢 @xuhuihuangchem.bsky.social and colleagues present MEMnets, combining statistical mechanics theory with DL to find the slowest collective variables for biomolecular dynamics. @uwdsi.bsky.social @uwmadscience.bsky.social #chemsky www.nature.com/articles/s43...

🔓https://rdcu.be/eqnCf

A committor-based method to uniformly sample rare reactive events - Nature Computational Science Enhanced sampling methods aim to simulate rare physical and chemical reactive processes involving transitions between long-lived states. Existing methods often disproportionally sample either metastab...

An accompanying Research Briefing is now available for this paper! #chemsky www.nature.com/articles/s43...

🔓https://rdcu.be/epJ6B

Leveraging large language models for pandemic preparedness - Nature Computational Science A framework with large language models is proposed to predict disease spread in real-time by incorporating complex, multi-modal information and using a artificial intelligence–human cooperative prompt...

An accompanying News & Views by Narendra Dixit is also available for this paper! #episky 😷💉 www.nature.com/articles/s43...

🔓https://rdcu.be/epJ0g

Advancing real-time infectious disease forecasting using large language models - Nature Computational Science PandemicLLM adapts the large language model to predict disease trends by converting diverse disease-relevant data into text. It responds to new variants in real time, offering robust, interpretable fo...

📢Yiran Chen, Lauren M. Gardner, Hao ‘Frank’ Yang, and colleagues introduce a framework that adapts LLMs to predict disease trends, offering forecasts for effective public health responses. #episky 😷💉 www.nature.com/articles/s43...

🔓https://rdcu.be/epJZk

Unified deep learning framework for many-body quantum chemistry via Green’s functions - Nature Computational Science A data-efficient deep learning model developed to predict ground-state and photophysical properties of molecules and nanomaterials by learning many-body Green’s functions achieves an accuracy surpassi...

📢Tianyu Zhu and colleagues develop a DL model to predict ground-state and photophysical properties of molecules and nanomaterials, achieving beyond-DFT accuracy with high data efficiency. #chemsky www.nature.com/articles/s43...

🔓https://rdcu.be/eptkB

How molecular changes impact brain states and whole-brain activity: a multiscale approach - Nature Computational Science Predicting how molecular changes affect brain activity is a challenge in neuroscience. We introduced a multiscale modeling approach to simulate these microscopic changes and how they impact macroscale...

A Research Briefing is now available for this paper! #compneurosky #Neuroskyence www.nature.com/articles/s43...

🔓https://rdcu.be/ephaB

A computational approach to evaluate how molecular mechanisms impact large-scale brain activity - Nature Computational Science A computational framework is introduced to study how molecular changes impact brain activity, using biophysically grounded mean-field models to evaluate how drugs acting on synaptic receptors lead to ...

📢Alain Destexhe and colleagues from @neuropsi.bsky.social introduce a computational framework to study how molecular changes impact brain activity. #compneurosky #Neuroskyence www.nature.com/articles/s43...

Abstract polygonal, triangular pale pink, light blue, and peach color background.

🚨Our May issue is now live and includes Perspectives on crystalline materials design and probabilistic photonic computing, a benchmark for quantum computing software, and much more! www.nature.com/natcomputsci...

When disruption endures - Nature Computational Science A new framework disentangles the nature of disruption in science, revealing how rare but persistent breakthroughs shake the foundations of research fields while remaining central to future work.

📢Russell Funk and Xiangting Wu discuss a recent framework that characterizes innovations that both disrupt earlier works and serve as stable foundations for subsequent development. #cssky #NetSci www.nature.com/articles/s43...

🔓https://rdcu.be/en1h4
📄https://rdcu.be/en1iS

Algorithms for reliable decision-making need causal reasoning - Nature Computational Science Decision-making inherently involves cause–effect relationships that introduce causal challenges. We argue that reliable algorithms for decision-making need to build upon causal reasoning. Addressing t...

📢Christoph Kern and colleagues argue in a Comment that reliable algorithms for decision-making need to build upon causal reasoning. #CausalSky www.nature.com/articles/s43...

🔓https://rdcu.be/enYdx

Interpretable niche-based cell‒cell communication inference using multi-view graph neural networks - Nature Computational Science An interpretable tool called STCase is introduced to leverage a multi-view graph neural network based on cell‒cell communication (CCC)-aware attention to uncover functional niche-specific CCC events b...

📢Wanqiu Ding and colleagues present STCase, a tool for uncovering functional niche-specific cell‒cell communication events based on spatial transcriptomics data. 🖥️ 🧬 www.nature.com/articles/s43...

🔓https://rdcu.be/enX9T