Can masked language modeling for molecules be improved? Fabian Krüger aichemist.eu/fellows investigates this question at doi.org/10.1039/D5DD... Using higher masking ratios, optimal pre-training dataset & model size provides better performance. Use the model from Hugging Face & enjoy better results

EU-OPENSCREEN and SLAS Launch the Second Joint Machine Learning Challenge EU-OPENSCREEN and the Society for Laboratory Automation and Screening (SLAS) are pleased to announce the second EU-OPENSCREEN/SLAS Joint Machine Learning Challenge, inviting scientists worldwide to pa...

Join the Second Joint Machine Learning Challenge to predict the optical properties of small molecules, transmittance and fluorescence, using screening data for 100k compounds Two winning teams will each receive a €1k prize during SLAS2026. Join ochem.eu/static/chall... & submit models by 15 Jan 2026

Visit of Prof. Yan ai-dd.eu/aixia partner of @aidd.bsky.social from Beijing University of Chemical Technology is always a big pleasure! After the lecture, I learned about recent student projects, had some delicious lunch and dinner at local restaurants, and also took a pleasant walk around Beijing!

Earlier this month, Lithuania hosted the 34th International Conference on Artificial Neural Networks – a major international event in artificial intelligence and neuroscience organised by ENNS, EBRAINS, LSMU, and VMU.

Read more: www.ebrains.eu/news-and-eve...

I was pleased to participate in Vincenzo Palmacci's @aidd.bsky.social PhD defence ai-dd.eu/vincenzo He presented an interesting paper on computational strategies for combating assay interference in high-throughput screening (HTS). Congratulations to Vincenzo and best wishes for your postdoc at NIH!

🧠💊 The 2nd AIDD Workshop kicks off next week at #ICANN2025 in Kaunas! Brought to you by AiChemist MSCA-DN!

Join top minds in AI + drug discovery for cutting-edge talks on XAI, molecular modeling & more.

Details 👉 aichemist.eu/icann2025
#AI #DrugDiscovery #Chemoinformatics

Advancing Human and Environmental Safety Science Using In Silico Methods

A Special Issue of #ChemResTox doi.org/10.1021/acs.... is calling for contributions based on accurate, validated, and regulatory-applicable approaches, in particular using mechanistic explanations (e.g., linking to MIE/AOPs) + traditional & new areas like big data, LLMs, XAI. Submit before 31/12!

Advanced machine learning for innovative drug discovery - Journal of Cheminformatics This editorial presents an analysis of the articles published in the Journal of Cheminformatics Special Issue “AI in Drug Discovery”. We review how novel machine learning developments are enhancing st...

Special Issue "AI in Drug Discovery" highlights how advanced machine learning enhances structural-based drug discovery, molecular property forecasting, and chemical reaction prediction. Enjoy reading the editorial rdcu.be/ezXFl as well as access all articles at www.biomedcentral.com/collections/...

VitroBert: modeling DILI by pretraining BERT on in vitro data - Journal of Cheminformatics Drug-induced liver injury (DILI) presents a significant challenge due to its complexity, small datasets, and severe class imbalance. While unsupervised pretraining is a common approach to learn molecu...

A VitroBERT, a molecular representation framework that integrates large-scale in vitro assays during pretraining, injecting biological context into molecular representations to improve modeling drug-induced liver injury was published in jcheminf.biomedcentral.com/articles/10....! Enjoy reading!

Igor Tetko gave an on-line seminar saferworldbydesign.com/webinars/56/ with an overview of #Tox24 challenge (see pre-print with S.A. Eytcheson doi.org/10.26434/che...)
The advantages of consensus models, which contributed most of winning models, were discussed. See also linkedin.com/feed/update/...

It was a pleasure to overview winning strategies to build machine learning models during the Erasmus Mundus Summer School on Chemoinformatics molekule.net/css2025/ at Ljubljana. Many thanks organisers for a great scientific and cultural program and interesting interactions with students and speakers

Are you curious about what protein language models learn? Check out our newest preprint! 🚀https://arxiv.org/abs/2506.19532
We reviewed explainable AI (XAI) techniques across all parts of the generative protein design workflow and discussed their applications, limitations, and untapped potential!

Consensus Modeling Strategies for Predicting Transthyretin Binding Affinity from Tox24 Challenge Data Transthyretin (TTR) is a key transporter of the thyroid hormone thyroxine, and chemicals that bind to TTR, displacing the hormone, can disrupt the endocrine system, even at low concentrations. This st...

Congratulations Thalita Cirino, @m-iwan.bsky.social from @aichemist.bsky.social and all their coauthors with publication pubs.acs.org/doi/10.1021/... summarising 9 models contributing top predictions of Transthyretin Binding Affinity in #Tox24 Challenge, which is published by the ChemResTox today!

Congratulation @m-iwan.bsky.social aichemist.eu/fellows with the Best Oral presentation award during 21st International Workshop on Quantitative Structure-Activity Relationships in Environmental and Health Sciences (QSAR2025) qsar2025.marionegri.it Looking forward towards your publication Mateusz!

Congratulations to Paula Torren Peraire www.linkedin.com/in/paula-tor... with excellent defense of her doctoral dissertation with "Summa cum laude" - the highest academic honor from @tum.de. All the best Paula with your future scientific career!

Helmholtz Munich @www.helmholtz-munich.de has most important event: evaluation of Project Oriented Funding (POF) which will determine future and funding of the centre for next 7y. Igor Tetko is presenting activities within @aidd.bsky.social & @aichemist.bsky.social in new beautiful congress centre.

Enhancing Transthyretin Binding Affinity Prediction with a Consensus Model: Insights from the Tox24 Challenge Transthyretin (TTR) plays a vital role in thyroid hormone transport and homeostasis in both the blood and target tissues. Interactions between exogenous compounds and TTR can disrupt the function of t...

The second article describing group winning model of #Tox24 challenge co-organised with @aidd.bsky.social was just published by @pubs.acs.org pubs.acs.org/doi/10.1021/... Congratulations to Xiaolin Pan @xlpan.bsky.social and his co-authors! Do not miss reading about strategies how to win Challenges!

Promotional graphic featuring the table of contents image along with the journal and article title.

'Machine Learning for Toxicity Prediction Using Chemical Structures: Pillars for Success in the Real World' from Chemical Research in Toxicology is an open access #ACSEditorsChoice.

📖 Read the article: buff.ly/61QkJhk

Sincere congratulations to Peter Hartog ai-dd.eu/peter with award of Dr. rer. nat. from @tum.de. Many thanks @fabiantheis.bsky.social & @itisalist.bsky.social for supervision and being members of commision & Samuel Genheden for co-supervising Peter at @astrazeneca.bsky.social All best wishes Peter!

The AiChemist Spring School in Lausanne is kicking off with a lecture on atom-centred representations for atomistic machine learning by Davide Tisi from @materials-epfl.bsky.social.

Igor Tetko summarized Tox24 chemrxiv.org/engage/chemr... to Prof. Xia students (Institute of Materia Medica, Chinese Academy of Medical Sciences & Peking Union Medical College) and attended lecture of Prof. Wang scholar.google.de/citations?us..., his former colleague from @www.helmholtz-munich.de

Igor Tetko lecture "Automated multistep retro-synthesis: current state and challenges" overview @aidd.bsky.social fellows' results gzs.buct.edu.cn/2025/0321/c5... to students of project partner Prof. Yan ai-dd.eu/aixia at Beijing University of Chemical Technology, followed by the hospitality lunch.

What a fantastic line-up!

But wait, there's more! 💫 Check out the full speaker list at www.cecam.org/workshop-det....

Registration is open until the 28th of March 📅 Don't miss out!

📢 ICANN 2025 – Second Call for Papers!

🚀 Deadline extended to March 29, 2025

📍 Kaunas, Lithuania | Sept 9-12, 2025
🔬 Topics: AI, Large Language Models, Brain-Inspired Computing, Robotics, and more

🌐 Learn more: e-nns.org/icann2025/ca...

#ICANN2025 #AI #ML #NeuralNetworks #DeepLearning #ENNS

The 2nd Workshop on AI in Drug Discovery (e-nns.org/icann2025/aidd) to be held within the 34th International Conference on Artificial Neural Networks (ICANN 2025) by @e-nns.bsky.social, invites cutting-edge contributions in the AI-driven drug discovery. Submit your article/abstract before 15 April!

Participate in the 34th International Conference on Artificial Neural Networks from 9-12 September in Kaunas, Lithuania!

Do you have a proposal for a special session or workshop at the event? Make sure to send it in before 1 March!

Learn more: www.ebrains.eu/news-and-eve...

34th International Conference on Artificial Neural Networks ICANN 2025

Submission system is now open.

Contribute to ICANN 2025 hosted in Kaunas, Lithuanian (Sep 9 - 12, 2025)

Submission deadline: March 15, 2025

See: e-nns.org/icann2025/su...

#ENNS #AI #NeuralNetworks #ML #MachineLearning

The #Tox24 Challenge was co-organized by @e-nns.bsky.social, @aidd.bsky.social and ChemResTox @pubs.acs.org. More models will be published soon. Stay tuned!

Consensus Modeling for Predicting Chemical Binding to Transthyretin as the Winning Solution of the Tox24 Challenge The utilization of predictive methodologies for the assessment of toxicological properties represents an alternative approach that facilitates the identification of safe compounds while concurrently reducing the financial costs associated with the process. The objective of the Tox24 Challenge was to assess the progress in computational methods for predicting the activity of chemical binding to transthyretin (TTR). In order to fulfill the requirements of this task, the data set, measured by the Environmental Protection Agency, consisted of 1512 chemical substances of diverse nature. This paper describes the model that won the Tox24 Challenge and the steps taken for its further improvement. The Transformer convolutional neural network (CNN) model achieved the best performance as a standalone solution. Meanwhile, a multitask model built on a graph CNN, trained using 11 additional acute systemic toxicity data sets with increased weighting on the TTR binding activity, showed comparable results on the blind test set. The winning solution was a consensus model consisting of two catBoost models with OEstate and Mold2 descriptor sets, as well as two transformer-based models. The improvement of this solution involved adding a fifth model based on multitask learning using the graph CNN method, which led to a reduction in RMSE on the blind test set to 20.3%. The winning model was developed using the OCHEM web platform and is available online at https://ochem.eu/article/162082.

The model that won the #Tox24 Challenge e-nns.org/icann2024/ch... has just been published by the ACS doi.org/10.1021/acs.....

The consensus model used representation learning doi.org/10.1186/s133... and mixture descriptors. Check out doi.org/10.26434/che... for an overview of the other top models.

Investigations into the Efficiency of Computer-Aided Synthesis Planning The efficiency of machine learning (ML) models is crucial to minimize inference times and reduce the carbon footprints of models deployed in production environments. Current models employed in retrosy...

Hartog et al. investigate ways to speed up the Chemformer retrosynthesis model with knowledge distillation, alternate architectures, and hyper-parameter tuning. pubs.acs.org/doi/full/10....