Sign in

Filip Miljković

@filipmiljkovic.bsky.social
14 followers 8 following 15 posts

Associate Principal AI Scientist @ AstraZeneca | Visiting Researcher @ Uppsala University | Affiliate Scientist @ University of Bonn | filipm90.github.io

PostsRepliesMedia
Filip Miljković @filipmiljkovic.bsky.social · 05/07/2026
Thanks for featuring our work, @iwatobipen.bsky.social! @edgarlopez2.bsky.social @jose-difacquim.bsky.social
010
Filip Miljković @filipmiljkovic.bsky.social · 21/06/2026
Our paper describing curation of bioactivity data for drugs and their associated metabolites is now published in Molecular Pharmaceutics. We were surprised to see that 26% of metabolites had either increased or retained activity compared to their parent drugs. Paper: pubs.acs.org/doi/10.1021/...
221
Filip Miljković @filipmiljkovic.bsky.social · 15/06/2026
Our advanced molecular fragmentation algorithm CRUSH is now published in Journal of Cheminformatics. Check it out here: link.springer.com/article/10.1... with @edgarlopez2.bsky.social, @jose-difacquim.bsky.social
link.springer.com
CRUSH—Cleavage Rules Using SMIRKS Heuristics: an enhanced molecular fragmentation algorithm - Journal of Cheminformatics
Molecular fragmentation methods are central to fragment-based discovery, virtual screening, and de novo molecular design, yet most existing approaches remain constrained by classical retrosynthetic ru...
130
Reposted by Filip Miljković
Rocío Mercado Oropeza @rociomer.bsky.social · 02/03/2026
Excited to share our pre-print on the curation of a new bioactivity dataset for metabolic transformations. 🧪🧑‍💻 We were surprised to find that roughly a quarter of our drug-metabolite-target combinations contain metabolites with retained or increased bioactivity relative to the parent drugs! #chemsky
1103
Reposted by Filip Miljković
Rocío Mercado Oropeza @rociomer.bsky.social · 02/03/2026
Work led by industrial PhD student Sofia Larsson and in close collaboration with Susanne Winiwarter and @filipmiljkovic.bsky.social at AstraZeneca. 📄 Paper: chemrxiv.org/doi/full/10.... 💻 Code: github.com/tsofiac/drug... 💽 Data: zenodo.org/records/1874...
chemrxiv.org
Systematic Computational Identification of Active Drug Metabolites Across the Human Proteome | ChemRxiv
Understanding drug metabolism helps mitigate toxicity risks and anticipate pharmacological effects beyond the parent compound. Here, we explore high-confidence in vitro human bioactivity annotations from the public domain associated with a curated dataset ...
183
Reposted by Filip Miljković
José L. Medina-Franco @jose-difacquim.bsky.social · 15/02/2026
The Issue “Good Practices in AI: Democratization of Open Science” is out. Honored to serve as Guest Editor and work with outstanding colleagues @filipmiljkovic.bsky.social @rociomer.bsky.social @wpwalters.bsky.social 👉 www.sciencedirect.com/special-issu... #AI #OpenScience #Chemoinformatics
042
Reposted by Filip Miljković
Rocío Mercado Oropeza @rociomer.bsky.social · 30/10/2025
Excited to share our recent paper, ”Compressing Biology,” to be presented at the Imageomics workshop at NeurIPS 2025. 🔬💻 Work led by my amazing PhD student Télio Cropsal. #cellpainting #stablediffusion #imageomics arxiv.org/abs/2510.19887
arxiv.org
Compressing Biology: Evaluating the Stable Diffusion VAE for Phenotypic Drug Discovery
High-throughput phenotypic screens generate vast microscopy image datasets that push the limits of generative models due to their large dimensionality. Despite the growing popularity of general-purpos...
163
Filip Miljković @filipmiljkovic.bsky.social · 22/10/2025
And Part II of our Metabolite Identification (MetID) Data in Drug Discovery is out: pubs.acs.org/doi/10.1021/... Here we explore application of MetID proprietary data to expand the capacity of site-of-metabolism (SoM) #ML models. The paper and SoM data are both #openacess. #Zenodo #ADME
pubs.acs.org
Metabolite Identification Data in Drug Discovery, Part 2: Site-of-Metabolism Annotation, Analysis, and Exploration for Machine Learning
The ability to pinpoint and predict sites of metabolism (SoMs) is essential for designing and optimizing effective and safe bioactive small molecules. However, the number of molecules with annotated S...
010
Filip Miljković @filipmiljkovic.bsky.social · 16/10/2025
Our paper on metabolite identification (MetID) data generation and trend analysis is now live and #openacess in Mol. Pharmaceutics @acs.org. In addition, we share proprietary MetID schemes for 120 compounds on #Zenodo. Link: pubs.acs.org/doi/10.1021/... Stay tuned for Part II! #dmpk #openscience
pubs.acs.org
Metabolite Identification Data in Drug Discovery, Part 1: Data Generation and Trend Analysis
In drug discovery, metabolite identification data are used to identify metabolic soft spots in research molecules to facilitate reduced metabolism in subsequently designed compounds. In addition, know...
011
Filip Miljković @filipmiljkovic.bsky.social · 22/09/2025
Pleased to announce that our research on LAGOM, a transformer-based model for predicting chemical structures of potential drug metabolites, has been published in Artificial Intelligence in the Life Sciences. #compchem #chemsky
030
Reposted by Filip Miljković
Rocío Mercado Oropeza @rociomer.bsky.social · 22/09/2025
Excited to share that our paper "LAGOM: A transformer-based chemical language model for drug metabolite prediction" has been accepted in AILSCI! doi.org/10.1016/j.ai... Work led by Sofia Larsson and Miranda Carlsson, with @rbeckmann.bsky.social and Filip Miljković‬ (AZ)! #compchem #chemsky
3186