Lukas Friedrich (@friedo91) profile picture

Lukas Friedrich

@friedo91

Data Scientist in Computational Chemistry at Merck Healthcare (Merck KGaA). Opinions are my own and not the views of my employer.

 Darmstadt, Deutschland
Joined February 2012
  • Tweets354
  • Following380
  • Followers350
22Media
Photo by Lukas Friedrich (@friedo91)Photo by Lukas Friedrich (@friedo91)Photo by Lukas Friedrich (@friedo91)Photo by Lukas Friedrich (@friedo91)Photo by Lukas Friedrich (@friedo91)Photo by Lukas Friedrich (@friedo91)Photo by Lukas Friedrich (@friedo91)Photo by Lukas Friedrich (@friedo91)Photo by Lukas Friedrich (@friedo91)Photo by Lukas Friedrich (@friedo91)
🚨 #Job #DeepTech We are actively looking for a #MachineLearning #Scientist with expertise in developing and using data-driven models, numerical #CompChem proficiency, and strong coding skills. Reach out now: quastify.com/jobs/ #rt appreciated
Image from Gerit Brandenburg (@JGBrandenburg)
Image from Gerit Brandenburg (@JGBrandenburg)
2
26
1
49
14,523
Will be around until Thursday if you have missed my talk. #ACSSpring2024
Johannes Schimunek (@JSchimunek) profile pictureJohannes Schimunek@JSchimunek18 Mar 2024
🚨 Live demo at ACS Spring 2024 🚨 For the lucky ones participating #ACS2024, check out the live demo by @friedo91 ⏰ 08:30am - 09:00am CDT - March 18, 2024 📌 Location: Hall B, Room 3 (Ernest N. Morial Convention Center)
1
6
728
🚨 Live demo at ACS Spring 2024 🚨 For the lucky ones participating #ACS2024, check out the live demo by @friedo91 ⏰ 08:30am - 09:00am CDT - March 18, 2024 📌 Location: Hall B, Room 3 (Ernest N. Morial Convention Center)
Johannes Schimunek (@JSchimunek) profile pictureJohannes Schimunek@JSchimunek18 Mar 2024
📢 Interested in activity predictions 🎯 for low data scenarios in drug discovery❓ 🚀 Use our app and prompt ⌨️ our model by providing just a few active 🟢​ and inactive 🔴 molecules! 🤗 Link: huggingface.co/spaces/ml-jku… 🚨 No programming skills needed!
Image from Johannes Schimunek (@JSchimunek)
1
1
4
919
Finally I made it. Looking forward to day 2 of #AIChem23 in Cambridge, UK.
1
13
799
Lukas Friedrich retweeted
Günter Klambauer (@gklambauer) profile picture
Günter Klambauer@gklambauer1 Sep 2023
Predicting molecular activity with XGBoost. 📊🧪 Study on feature importance, highlighting the need for expert interpretation. Hyperparameter optimization is crucial. Valuable guidelines for #cheminformatics practitioners. #ML Paper: link.springer.com/article/10…
Image from Günter Klambauer (@gklambauer)
7
60
3,088
Lukas Friedrich retweeted
ChemMedChem (@ChemMedChem) profile picture
ChemMedChem@ChemMedChem16 Jun 2023
Metabolism of (R)-Praziquantel versus the Activation of a Parasite Transient Receptor Potential Melastatin Ion Channel (Marchant, Spangenberg et al.) @MCW_CBNA @MerckHealthcare @ISNTD_Press onlinelibrary.wiley.com/doi/…
Image from ChemMedChem (@ChemMedChem)
3
6
1,019
Code availability: github.com/ml-jku/MHNfs
Image from Johannes Schimunek (@JSchimunek)
2
3
236
We benchmark our model on FS-Mol, provide an ablation study, and perform a domain-shift experiment.
1
1
3
216
🚀 Excited to share our #ICLR2023 work on 🚨 context-enriched molecule representations🚦 improve few-shot drug discovery 💊 🚨 Paper: openreview.net/pdf?id=XrMWUu… App: HuggingFace 🤗 under prep! #ICLR2023 🧑‍💼 poster 🗨: iclr.cc/virtual/2023/poster/… ⏰ Wed 3 May 4:30 pm - 6:30 pm CAT
Image from Johannes Schimunek (@JSchimunek)
2
21
2
41
6,493
Lukas Friedrich retweeted
Ingo Hartung (@HartungIngo) profile picture
Ingo Hartung@HartungIngo13 Mar 2023
We have a job opening in our MedChem department in Darmstadt/GER (Frankfurt area): Join us to discover the next generation of drugs in oncology & immunology (small molecule inhibitors & degraders). Candidates with hands-on #MedChem experience preferred. merckgroup.com/en/careers/jo…
Image from Ingo Hartung (@HartungIngo)
Image from Ingo Hartung (@HartungIngo)
Image from Ingo Hartung (@HartungIngo)
1
16
3
39
13,044
Lukas Friedrich retweeted
Jonas Boström (@DrBostrom) profile picture
Jonas Boström@DrBostrom1 Mar 2023
A timely paper by Tim Cernak showing how ChatGPT can automatically formulate reaction arrays for common reactions chemrxiv.org/engage/chemrxiv… chemrxiv.org/engage/api-gate…

Designing Chemical Reaction Arrays using phactor and ChatGPT | ChemRxiv

High throughput experimentation (HTE) is a common practice in the pharmaceutical industry. Medicinal chemists design reaction arrays to optimize the yield of couplings between building blocks and/or...

chemrxiv.org
3
6
748
Lukas Friedrich retweeted
Julien Lefranc (@LefrancJu) profile picture
Julien Lefranc@LefrancJu16 Nov 2022
If your are interested PI3K inhibitors or just want to learn more about the field, check our article on @drughuntersite ! and let us know what you think
drug hunter (@drughunter_com) profile picturedrug hunter
@drughunter_com15 Nov 2022
Small Molecule PI3K Inhibition in Oncology: What's Been Done and What's to Come? | drughunters.com/3TBF6Gi We reviewed the history and current status of PI3K inhibitors, including their approval history, resistance mechanisms, and the next generation of PI3K Inhibitors
Image from drug hunter (@drughunter_com)
1
3
The results of our joint work with the 2020 #merckfutureinsightprize winner @SieberLab is online. Thanks to Davide and Stephan for your great work and a wonderful collaboration.#DrugDiscovery #MachineLearning @robofresh @MerckHealthcare @UlrichBetz
Journal of Cheminformatics (@jcheminf) profile pictureJournal of Cheminformatics@jcheminf14 Nov 2022
new: "Tuning gradient boosting for imbalanced bioassay modelling with custom loss functions" jcheminf.biomedcentral.com/a…
Fig. 3: Benchmarking results for the proprietary HTS dataset.

ALT Fig. 3: Benchmarking results for the proprietary HTS dataset.

1
5
19
Lukas Friedrich retweeted
Günter Klambauer (@gklambauer) profile picture
Günter Klambauer@gklambauer9 Nov 2022
ELLIS ML4Molecules 2022: will take place on Nov 28; FREE EVENT! Please register here: eventbrite.com/e/ellis-machi…
ELLIS Machine Learning for Molecule Discovery Workshop

ELLIS Machine Learning for Molecule Discovery Workshop

Big successes of machine learning (ML) for molecules have been achieved recently, e.g. the accurate prediction of protein 3D structure (Jump

eventbrite.com
18
2
48
Congratulations to all the @MELLODDY_IMI team! The manuscript titled "MELLODDY: cross pharma federated learning at unprecedented scale unlocks benefits in QSAR without compromising proprietary information" is now publicly available on ChemRxiv. chemrxiv.org/engage/chemrxiv… #MELLODDY

MELLODDY: cross pharma federated learning at unprecedented scale unlocks benefits in QSAR without...

Federated multi-partner machine learning can be an appealing and efficient method to increase the effective training data volume and thereby the predictivity of models, particularly when the genera...

chemrxiv.org
2
7
Here we go again. See you tomorrow in Berlin, #RdkitUGM2022
Image from Lukas Friedrich (@friedo91)
1
20
Lukas Friedrich retweeted
Claas Klar (@ClaaasKlar) profile picture
Claas Klar@ClaaasKlar5 Oct 2022
ÄBBELWOI NIEMALS MIT COLA!
Image from Claas Klar (@ClaaasKlar)
5
16
2
418
Lukas Friedrich retweeted
Günter Klambauer (@gklambauer) profile picture
Günter Klambauer@gklambauer12 Sep 2022
The ELLIS ML4Molecules workshop will also happen this year on November 28 in VIRTUAL format! Please find the announcement and the call for papers here: moleculediscovery.github.io/… Looking forward to your contributions!
Image from Günter Klambauer (@gklambauer)
2
79
6
212
We are thrilled to announce the final results of the @MELLODDY_IMI #IMI_JU project. We have demonstrated that collaborating in AI for drug discovery is possible and the results indicate potential benefits for the discovery of new drugs. Learn more at melloddy.eu/y3announcement
Image from MELLODDY Consortium (@MELLODDY_IMI)
1
4
1
12
Lukas Friedrich retweeted
Owkin (@OwkinScience) profile picture
Owkin
@OwkinScience13 Jul 2022
🚨@MELLODDY_IMI, a landmark project that enabled 10 pharma companies and 7 strategic partners to work together, has successfully shown that collaborating in AI for drug discovery is possible, indicating potential benefits for the discovery of new drugs. melloddy.eu/y3announcement
Image from Owkin (@OwkinScience)
1
8
1
13
Load more