News | Advanced machine learning for Innovative Drug Discovery (AIDD)

2024

06.12 Igor Tetko gave an invited lecture with an overview of AIDD results during Machine Learning for Molecules workshop [ML4Molecules]

28.11 Julian Cremer will defend his PhD at Universitat Pompeu Fabra on the 28th November at 11:00 CET. Best of luck to him!

23.10 Alan Kai Hassen gave a talk based on his soon-to-be published article "Generate what you can make: achieving in-house synthesizability with readily available resources in de novo drug design" at the October AIDD/AiChemist seminar.

17.10 Varvara Voinarovska successfully defended her PhD thesis at the Technical University of Munich. Congratulations Dr. Voinarovska!

17.10 "Low concentration cell painting images enable the identification of highly potent compounds" by Son Ha is published in Scientific Reports.

20.09 Julian Cremer was awarded "Best Student Paper Award" at ICANN2024. Congratulations!

20.09 The proceedings from the "AI in Drug Discovery" Workshop at ICANN2024 are published! The submissions from the AIDD consortium are published in the fully open-access volume of the ICANN2024 proceedings. The remainder are published in Part X of the proceedings.

19.09 The AIDD fellows attended the "AI in Drug Discovery" Workshop at ICANN2024, with many of them presenting their work in-person during the Workshop.

03.09 Igor Tetko gave a lecture "Advanced Machine Learning in Drug Discovery" at KU Leuven.

30.08 Mariia Radaeva successfully defended her PhD thesis. Congratulations Dr. Radaeva!

18.06 Dr. Iris Köhler and Dr. Sarah de Carvalho Hartmann of The Scientist Coach delivered workshops on "Competency Profiles" and "Job Applications" for the ESRs.

27.05 "Statistical approaches enabling technology-specific assay interference prediction from large screening data sets" by Vincenzo Palmacci is published as a pre-proof in AILSCI.

23.05 The pre-print "PILOT: Equivariant diffusion for pocket conditioned de novo ligand generation with multi-objective guidance via importance sampling" by Julian Cremer is online!

23.05 Together with the European Neural Network Society, AiChemist MSCA DN, and Chem. Rex. Toxicol., AIDD is delighted to announce the Tox24 Challenge! This challenge is designed to assess the state of the art in computational methods for predicting the in vitro activity of compounds, using a dataset comprising compounds tested for activity against transthyretin, obtained by the US EPA. Participants will be competing for a prize of 1000 EUR, to be awarded to the developers of the winning model. A description, with rules and timeline, can be found at https://ochem.eu/static/challenge.do

30.04 Vincenzo Palmacci gave a talk titled "Statistical approach enabling technology-specific assay interference prediction from large screening data sets".

15.04 "Tackling assay interference associated with small molecules" by Vincenzo Palmacci and coauthors has been published in Nature Reviews Chemistry

04.04 "Using test-time augmentation to investigate explainable AI: inconsistencies between method, model and human intuition" by Peter Hartog is published in Journal of Cheminformatics.

02.04 Emma Svensson gave a talk "HyperPCM: Robust Task-Conditioned Modeling of Drug-Target Interactions" for the M2D2 talk series from Valence Labs at Mila (April 2nd, 2024). Watch the talk here!

05.03 A pre-print of "Generate What You Can Make: Achieving in-house synthesizability with readily available resources in de novo drug design" by Alan Kai Hassen and coauthors has been deposited online

04.03 The final AIDD School will be taking place in Berlin, from 04.03 til 13.03. Lectures and workshops will be run in hybrid mode where possible. The agenda and Zoom link/s for the lectures can be found on the Lectures page.

21.02 A pre-print of "Alphafold meets de novo drug design: leveraging structural protein information in multi-target molecular generative models" by Alan Kai Hassen has been published online.

21.02 A pre-print of "Aquamarine: Quantum-Mechanical Exploration of Conformers and Solvent Effects in Large Drug-like Molecules" by Alessio Fallani and Mathias Hilfiker is published online

21.02 Mikhail Andronov gave a talk based on his recently uploaded pre-print, titled "A reagent-driven visual method for analyzing chemical reaction data"

15.02 "FSL-CP: a benchmark for small molecule activity few-shot prediction using cell microscopy images" by Son Ha has been published in Digital Discovery

15.02 AIDD will be co-organising a workshop (AI in Drug Discovery) at ICANN2024 together with the AiChemist project. The Call for Papers is open until 15.03 (15.04 for abstracts)

14.02 "Models Matter: the impact of single-step retrosynthesis on synthesis planning" by Paula Torren-Peraire and Alan Kai Hassen has been published in Digital Discovery.

08.02 The EPAA has published its call for the 2024 3Rs Student Grants, which will sponsor students to participate in the major 3Rs events in 2024. They will award a full grant of 1000€ and a half grant of 500€ for ESTIV 2024, EUSAAT 2024 and EUROTOX 2024. Email your applications to grow-epaa@ec.europa.eu by midnight on 15.04.24 for ESTIV and 17.06.24 for EUSAAT and EUROTOX. Good luck!

03.02 "The openOCHEM consensus model is the best-performing open-source predictive model in the First EUOS/SLAS Joint Compound Solubility Challenge" by Peter Hartog and co-authors is published in SLAS Discovery.

24.01 Pre-print of "A reagent-driven visual method for analyzing chemical reaction data" by Mikhail Andronov has been published online

07.01 "HyperPCM: Robust Task-Conditioned Modeling of Drug-Target Interactions" by Emma Svensson is published in Journal of Chemical Information and Modelling

2023

20.12 “When yield prediction does not yield prediction: an overview of the current challenges” is published in Journal of Chemical Information and Modelling

12.12 Igor Tetko gave a lecture, titled "OCHEM, openOCHEM and beyond: use of Advanced machIne learning (AI) in Chemistry" at PharmaCampus-Kolloquium, University of Münster (see summary)

12.12 Julian Cremer gave an online lecture, titled "Navigating the Design Space of Equivariant Diffusion-Based Generative Models for De Novo 3D Molecule Generation"

28.11 Arslan Masood gave an online lecture titled "Dissecting Drug-Induced-Liver-Injury"

13.11 CLOOME: contrastive learning unlocks bioimaging databases for queries with chemical structures by Ana Sanchez-Fernandez has been published in Nature Communications

25.10 Rosa Friesacher gave an online lecture, titled "Understanding Model Uncertainty and Enhancing Probability Calibration in Neural Networks"

23.10 "Modeling noncovalent interatomic interactions on a photonic quantum computer" by Alessio Fallani is published in Physical Review Research

20.09 Pre-print of "Navigating the design space of equivariant diffusion-based generative models for de novo 3D molecule generation" by Julian Cremer is published online

21.09 Pre-print of When yield prediction does not yield prediction: an overview of the current challenges by Varvara Voinarovska is published

19.09 Varvara Voinarovska gave an online lecture titled “When yield prediction does not yield prediction: an overview of the current challenges”

10.09 Equivariant Graph Neural Networks for Toxicity Prediction by Julian Cremer has been published in Chemical Research in Toxicology and got high recognition as

05.09 AiChemist project has started. Apply for 14 open positions!

21.08 Special Issue "AI meets Toxicology" is out at ChemResTox featuring 25 articles on cutting-edge machine learning applications, from classical methods to generativeAI to equivariant neural networks for improved estimation of human toxicity

13.08 Paula Torren gave a talk "Models Matter: The Impact of Single-Step Retrosynthesis on Synthesis Planning" at ACS meeting in San Francisco

10.08 Pre-print of "Models Matter: The Impact of Single-Step Retrosynthesis on Synthesis Planning" by Paula Torren and Alan Hassen is published

24.07 Igor Tetko gave lecture and provided practical demonstration "Development of QSAR/QSPR models using representation learning and descriptor based methods using openOCHEM. Which ones are better?" during OpenTox Summer 2023 School

19.07 Mariia Radaeva gave lecture "Targeting Androgen Receptor with CADD"

09.07 Igor Tetko gave on-line presentation "Science is fun but not only: career perspectives for young talented researchers in a modern society" with a focus on opportunities in European Union and Germany

07.07 During HMGU MTCC workshop a flash talk and a poster with Solubility prediction challenge results was presented

03-07.07 Fifth School of the project (open only to project partners) was organized at AstraZeneca

29.06 Igor Tetko gave on-line presentation and lecture "Introduction to Machine Learning and best practices for development and validation of QSPR/QSAR models using OCHEM"

22.06 Ana Messias gave a talk "Success stories of structure-based drug discovery"

19-20.06 Igor Tetko presented OCHEM during the Final Workshop of the EU LIFE CONCERT REACH project

15.06 Silvano Coletti gave a talk "Scale your Business idea"

26.05 Paula Torren gave a talk "Enhancing Chemical Synthesis Planning through Combining Single-Step and Multi-Step Retrosynthesis Prediction Strategies" at the seminar of Institute of Structural Biology

25.05 Invited lecture to present winning model of the Kaggle 1st EUOS/SLAS Joint Challenge: Compound Solubility Challenge at SLAS 2023 conference in Brussels by HMGU team

24.05 Mikhail Andronov gave a talk about reagent prediction at the DigiDrug seminar in Berlin organized by Prof. Andreas Bender

22.05 Paula Torren speaks at "Pint of Science Munich" - an event to promote science to general public, see more at twitter

17.05 On-line seminar of Son Ha FSL-CP: Few-shot Prediction of small molecule activity using cell microscopy images

12.05 Six Newsletter was published. You can subscribe to Newsletters here

05.04 An announcement for PM position has been made.

01.04 Our Project Manager, Marchela Pandelova got full-time permanent position at HMGU. We sincerely congratulate her and thank for nice work!

28.03 Reagent prediction with a molecular transformer improves reaction data quality by Mikhail Andronov and Varvara Voinarovska is published in Chemical Science

23.03 Grant Amendment was signed by EU. We welcome Pfizer and Chelonia as new partners of the project

20-23.03 Fourth AIDD School took place in Aalto university

15-17.03 AIDD fellows participated to BayesComp2023 conference at Levi

9.03 A team from HMGU won the 1st EUOS/SLAS Joint Challenge: Compound Solubility Challenge

8.03 On-line Seminar Alessio Fallani: "Extending 3D generative modeling of molecules with quantum-mechanical properties"

28.02 Fifth Newsletter was published. You can subscribe to Newsletters here

17.02 Lecture by Dr. Hyun Kil Shin "ToxSTAR for DILI prediction"

14.02 Article by Mikhail Andronov and Varvara Voinarovska is accepted in Chemical Science. Congratulations!

02.02 AIDD won a bid to organize ENNS ICANN2024 in IDSIA, Lugano, Switzerland to be co-chaired by Jürgen Schmidhuber and supported by other AIDD PIs

01.02 On-line Seminar by M. Andronov "Reagent prediction with a transformer and its benefits for reaction product prediction" and by J. Cremer "Equivariant Graph Neural Networks for Toxicity Prediction"

01.02 Mathias Hilfiker has started as new fellow at the University of Luxembourg. Welcome Mathias!

2022

02.12. Public lecture of Vincenzo Palmacci at Bologna University "Drug Discovery and Cheminformatics: discovering new drugs in the Big Data era"

02.12. Several AIDD participants joined Thirty-sixth Conference on Neural Information Processing Systems were they presented their works

29.11 Chelonia has joined AIDD as a new associated partner to promote dissemination and outreach activities

25.11 Fourth AIDD conference will take place in Finland, 15-22 March 2023 (place and program will be announced soon)

17-25.10. Third AIDD conference took place at nice and a peaceful University city of Löwen

10.10 Program of Third School at KUL is announced

21.07 OpenTox Summer School 2022 presentation of openOCHEM by Igor Tetko

15.07 Open source OCHEM is published at https://github.com/openochem

13.07 "A tour through molecular representations in AI-driven drug discovery" by Laurianne David

22.06 On-line Seminar "AIDD Codebase: a Framework for Model Integration, Collaboration and Sharing" by Emma Svensson and Peter Hartog (only to AIDD partners), 13:00

17.06 Fourth newsletter is on-line

08.06.2022 On-line Seminar "Contrastive Learning of Image and Structure-Based Representations in Drug Discovery" by Ana Sanchez-Fernandez

24.05.2022 First article with AIDD fellow Ana Sanchez-Fernandez is published in proceedings of ICLR2022 Machine Learning for Drug Discovery conference. Congratulations Ana!

19.05 Lecture "Latent-space Chemistry: Deep Learning for Drug Discovery" by Djork-Arné Clevert at "Simulation and AI in the Future of Science - Beilstein Bozen Symposium", Hotel Jagdschloss Niederwald - Rüdesheim, Germany

09.05-18.05 Summer School on "Advanced Machine Learning", Lugano, Switzerland

17.03 Alessio Fallani gave a talk "Towards the inverse design of molecules with targeted quantum-mechanical properties" at APS March Meeting 2022, https://meetings.aps.org/Meeting/MAR22/Session/T49.13

16.03. On-line Seminar on presentation of the paper "A survey of human-in-the-loop for machine learning" by Yasmine Nahal. The presentation is available here.

02.03 On-line Seminar "Computer-Aided Synthesis Planning Tools to Facilitate Experimental Engagement" by Dr. Amol Thakkar. The presentation is available here.

23.02 On-line Seminar "3D structure refinement" by Dr. Filipe Menezes. The presentation is available here.

27.01 On-line Seminar on presentation of the paper " Coarse graining molecular dynamics with graph neural networks" by Andrey Mossyayev. The presentation is available here.

2021

20.12 On-line Seminar. Samuel Genheden (AstraZeneca) gave a talk about AiZynthFinder. The talk is available on https://youtu.be/r9Dsxm-mcgA

09.12 The ESR3 position is filled.

08.12 On-line Seminar on presentation of the paper "Non-Autoregressive Electron Redistribution Modeling for Reaction Prediction" written by Hangrui Bi, Hengyi Wang, Chence Shi, Connor Coley, Jian Tang, Hongyu Guo. http://proceedings.mlr.press/v139/bi21a/bi21a.pdf Mikhail Andronov presented the paper and answered all the questions. All fellows participated.

09.11 ESR3 position is advertised on https://euraxess.ec.europa.eu/jobs/660545

05.11 The ESR3 position is reopened.

29.10 Hiking in Schliersee region. closing the School.

19.10-28.10 Lectures and workshop on HMGU Campus. Some of the lectures were given by Zoom.

18.10 The first School is started. Welcome party in Biergarten Chinesischen Turm in Muenchen.

07.10 The ESR1 position is filled

20.08 The ESR1 position is advertised on https://www.linkedin.com/posts/esbenbjerrum_phd-positions-activity-6834394790017216512-Agzj and https://twitter.com/ChemITnerf/status/1428630595492536321?s=20

18.08 The ESR1 position is reopened. Announcement is published on https://euraxess.ec.europa.eu/jobs/675339

17.08 The ESR3 position is filled.

07.07 The ESR3 position is published on https://euraxess.ec.europa.eu/jobs/660545

30.06 Reopening of the ESR3 position

30.06 General Assembly meeting

31.05 First Newsletter is published, subscribe page is available on the site.

25.05 Candidates are notified about the results of the interviews. 15 fellows were nominated to 15 ESR projects.

28.04 - 14.05 Recruitment meetings via a series of Zoom meetings are conducted.

18.04 Deadline for application submission, 260 submissions are received.

18.03 Positions are announced on Mendeley https://www.mendeley.com/careers/job/phd-positions-aidd-h2020-msca-itn-2020-project-development-interpretable-deep-neural-networks

17.03 Positions are announced on nature.com https://www.nature.com/naturecareers/job/15-phd-positions-development-of-one-chemistry-unified-and-interpretable-deep-neural-networks-model-for-drug-discovery-helmholtz-zentrum-munchen-german-research-center-for-environmental-health-hmgu-737900

17.03 Positions are announced on LinkedIn https://www.linkedin.com/jobs/view/2443203197

08.03 Positions are announced on LinkedIn http://www.linkedin.com/hiring/jobs/2452233083

02.03 Job offers on http://www.ccl.net/cca/jobs/joblist/mess0053939.shtml

24.02 ASCENION announces the the AIDD project

26.01 Kick-off meeting, introductory presentations of the Partners are below: