Lightnews — Scholar-powered news

DaiSyBio @daisybio.de · Jul 16

#teamretreat
Hello from the Bavarian Alps 👋
Team building activities, vision development and a hackathon are on our agenda: some days beyond the routine to rethink team work, to cook together, to listen to each other’s talks and all this surrounded by a splendid landscape!

2

Reposted by DaiSyBio

REPO4EU @repo4eu.bsky.social · Jun 10

REPO4EU partners in action 🦾 The @cosybio-uhh.bsky.social team recently hosted a workshop at @uni-hamburg.de to validate the AI-powered #DrugRepurposing workflows being developed by researchers from @daisybio.de, @univie.ac.at and @upm.es

Learn more about our mission 💊 repo4.eu/the-platform/

From left to right: Lisa Marie Spindler, Ana Casas, Johannes Kersting, Lorenza D'Alessandro, Fernando M. Delgado-Chaves, Markus List, Jan Baumbach, Chloé Bucheron, Cristian Nogales, Pablo Perdomo, Farzaneh Firoozbakht and Joaquim Aguirre-Plans.

REPO4EU researchers from Technical University of Munich, University of Vienna, Universidad Politécnica de Madrid and STALICLA together during a workshop organised by the CoSyBio research group at University of Hamburg.

2 2

Reposted by DaiSyBio

Markus List @itisalist.bsky.social · Jun 3

🧬🖥️Drug response prediction is a machine learning challenge with immense potential for precision medicine. Our latest preprint introduces DrEval, a comprehensive benchmarking framework to evaluate state-of-the-art methods, uncover widespread issues, and guide the development of more robust models.

Judith Bernett @judith-bernett.bsky.social · Jun 3

🧬🖥️So excited to show you the outcome of @pascivers.bsky.social and my latest project: "From Hype to Health Check: Critical Evaluation of Drug Response Prediction Models with DrEval" doi.org/10.1101/2025.05.26.655288, published with M. Picciani, M. Wilhelm, K. Baum & @itisalist.bsky.social.
🧵1/10

Overview of the DrEval framework. Via input options, implemented state-of-the-art models can be compared against baselines of varying complexity. We address obstacles to progress in the field at each point in our pipeline: Our framework is available on PyPI and nf-core and we follow FAIReR standards for optimal reproducibility. DrEval is easily extendable as demonstrated here with a pseudocode implementation of a proteomics-based random forest. Custom viability data can be preprocessed with CurveCurator, leading to more consistent data and metrics. DrEval supports five widely used datasets with application-aware train/test splits that enable detecting weak generalization. Models are free to use provided or custom cell line– and drug features. The pipeline supports randomization-based ablation studies and performs robust hyperparameter tuning for all models. Evaluation is conducted using meaningful, bias-resistant metrics to avoid inflated results from artifacts such as Simpson’s paradox. All results are compiled into an interactive HTML report. Created in https://BioRender.com.

5 9

Reposted by DaiSyBio

Judith Bernett @judith-bernett.bsky.social · Jun 3

🧬🖥️So excited to show you the outcome of @pascivers.bsky.social and my latest project: "From Hype to Health Check: Critical Evaluation of Drug Response Prediction Models with DrEval" doi.org/10.1101/2025.05.26.655288, published with M. Picciani, M. Wilhelm, K. Baum & @itisalist.bsky.social.
🧵1/10

1 3 16

Reposted by DaiSyBio

Markus List @itisalist.bsky.social · May 27

For those of you who are not in Innsbruck to see me today, you might instead listen to @judith-bernett.bsky.social at the @iscb.bsky.social NetBio webinar!

🔗 Attend at ISCB Nucleus: iscb.junolive.co

📍 If you’re not an ISCB member, register for access to ISCB Nucleus: lnkd.in/gMhrKGJz

1 2

DaiSyBio @daisybio.de · May 27

Today, Prof. Markus List @itisalist.bsky.social will be giving a talk at Universität Innsbruck: "Data leakage is a widespread issue in machine learning in the biomedical domain: How to spot it, avoid it, and fix it"
🎤 Hosted by Prof. Francesca Finotello @francescafinotello.bsky.social See you there!

1 1

Reposted by DaiSyBio

Tanja Laske @mathbiotanja.bsky.social · May 11

Check out our new paper in Briefings in Bioinformatics: #PRONE - an R package & app for evaluating a plethora of #proteomics normalization methods, showcased for nine datasets! Find out more here: doi.org/10.1093/bib/...
@cosybio-uhh.bsky.social @daisybio.de TUM FraunhoferIZI

Systematic evaluation of normalization approaches in tandem mass tag and label-free protein quantification data using PRONE

Abstract. Despite the significant progress in accuracy and reliability in mass spectrometry technology, as well as the development of strategies based on i

doi.org

1 5

Reposted by DaiSyBio

REPO4EU @repo4eu.bsky.social · Apr 16

🎙️ Welcome back to the REPO4EU Podcast!

On episode 2 you'll meet Johannes Kersting, Bioinformatician and PhD Researcher at @daisybio.de building a knowledge base for #DrugRepurposing.

YouTube 🔴 youtu.be/wQD8LbHEFYc?...

Spotify 🎧 open.spotify.com/episode/0yNj...

#PhDVoice #AcademicSky #Science

REPO4EU Podcast | Episode 2: Johannes Kersting (DaiSyBio / Technical University of Munich)

YouTube video by REPO4EU

youtu.be

1 2

Reposted by DaiSyBio

Alessandro Nicoli @anicoli90.bsky.social · Apr 4

🎙️Happy to have presented my #Postdoctoral project at the Weihenstephan #Bioinformatics Symposium 2025 organized by Prof. @itisalist.bsky.social and by the amazing team @daisybio.de at #TUM!!!

Learned a lot about the exciting science around #TUM ann #LMU 🤓

#compchem #MD #chemosensory #GPCR

1 9

DaiSyBio @daisybio.de · Apr 3

Weihenstephan Bioinformatics Symposium 2025: More than 75 scientists from Bavaria and the world came together to share talks and create new synergies. It was great to host this event, for those who missed it: The next edition is planned for 2027 😉

1 2 8

DaiSyBio @daisybio.de · Mar 20

Greetings from Palermo! @en-coding.bsky.social, @a-dietrich.bsky.social, @itisalist.bsky.social, Serafina Reif, Nico Trummer & Kamila Kwiecien are united here at the occasion of the MyeInfoBank COSTAction: Converting Molecular Profiles of Myeloid Cells into Biomarkers for Inflammation and Cancer

2 3

DaiSyBio @daisybio.de · Mar 12

📍#thisisus
@quirinmanz.bsky.social is today's doctoral candidate in focus! He is fascinated by the broad application of bioinformatics, from understanding the fundamentals of molecular biology to creating patient benefits in clinics. Currently, he is planning a research stay overseas. Stay posted!📍

5

DaiSyBio @daisybio.de · Mar 6

📍@daisybio.de #thisisus
@a-dietrich.bsky.social began his PhD in 2022. Alex is genuinely interested in how cells are defined by their molecular profiles and in cell-type deconvolution. Besides that, he is positioning DaiSyBio under the top participants at the TUM campus run! 🏃🎽👟 📍

6

DaiSyBio @daisybio.de · Feb 27

This week, we had the honour to have visitors from Japan! 🇯🇵
@rnakato.bsky.social Lab of Computational Genomics, University of Tokyo gave the seminar talk "Deep learning-based approaches to elucidate unknown functional regions of the genome". Looking forward to further collaboration opportunities!

1 7

Reposted by DaiSyBio

REPO4EU @repo4eu.bsky.social · Nov 19

Hello Bluesky! Let us introduce ourselves 👉 we're REPO4EU, a #EUfunded project made out of 28 partners working together to advance #DrugRepurposing for #PrecisionMedicine in Europe and beyond 💊 🇪🇺 🌍

If you're into #NetworkMedicine, #PharmaInnovation and #AIinHealthcare, let's connect! 🔄

🌐 repo4.eu

REPO4EU Consortium celebrating the 3rd General Assembly (Stockholm, November 2023).

3 12

DaiSyBio @daisybio.de · Feb 27

📍 @daisybio.de #thisisus
This is Annelore Hermann, our science manager. With a background in intercultural communication & international innovation management, Horizon Europe is her specialty. Making science accessible for citizens is essential to her as it spreads trust throughout society! 📍

4

DaiSyBio @daisybio.de · Feb 13

📍Welcome to our presentation round of the DaiSyBio members! Every week, you will get to know someone from our lab.
The start is done by @itisalist.bsky.social who heads the group. Markus joined TUM in 2018 and became a W2 tenure track associate professor in 2023. More members are about to follow! 📍

2 8

DaiSyBio @daisybio.de · Jan 30

Congrats to our PhD student Johannes Kersting for winning the ASAPbio poster prize for his contribution to the @repo4eu.bsky.social #REXPO24: blog.scienceopen.com/2025/01/joha...

Johannes Kersting Awarded an ASAPBio Poster Prize for RExPO Contribution

The ASAPBio poster competition is an initiative that champions open science by highlighting innovative research and the early sharing of results. This year’s competition emphasized transparency, colla...

blog.scienceopen.com

2 7

Reposted by DaiSyBio

Judith Bernett @judith-bernett.bsky.social · Jan 27

🧬🖥️ Proud to share our latest update on PPI predictions – "Deep learning models for unbiased sequence-based PPI prediction plateau at an accuracy of 0.65" doi.org/10.1101/2025... by T. Reim, published with @itisalist.bsky.social @dbblumenthal.bsky.social, A. Hartebrodt, and me. What did we do? 1/15 🧵

Graphical summary of the analyses done in the publication displayed on six panels a-f. (a) We computed ESM-2 embeddings of different sizes for the proteins of our data-leakage-free PPI dataset. The per-token embeddings have variable sizes depending on the protein length, while the per-protein embeddings have a fixed size by applying dimension-wise averaging. (b) We tested two models operating on the per-protein embeddings—a baseline random forest classifier and adaptions of the previously published Richoux model. Five models operated on the per-token embeddings: a 2d-baseline, the 2d-Selfattention and 2d-Crossattention models (which expanded the 2d-baseline through a Transformer encoder), and adaptations of the published models D-SCRIPT and TUnA. (c) Hyperparameter tuning gave us insight into the influence of each tunable parameter on the classification performance. (d) No model surpassed an accuracy of 0.65. The more advanced models had similar accuracies, leading us to believe that the information content of the ESM-2 embedding has more influence than the model architecture. Per-token models did not consistently outperform per-protein models. (e) We applied various modifications to test their influence: different embedding sizes, inserting a Transformer encoder into different positions, adding spectral normalization after the linear layers, self- vs. cross-attention, and removing the padding. (f) Finally, we compared the implicitly predicted distance maps of the 2d-baseline, 2d-Selfattention, 2dCrossattention, and D-SCRIPT-ESM-2 to real distance maps computed from PDB structures.

1 3 10

DaiSyBio @daisybio.de · Jan 23

📣 Shoutout to all young molecular biologists, bioinformaticians and immunobiologists: Don't miss our lab member Alexander Dietrich giving a talk on cell-type deconvolution next THU at 2pm 🕑
It's free, it's virtual, it's new! Register here ⤵️
bit.ly/3VI6Dtx

2

DaiSyBio @daisybio.de · Jan 8

Time for European news: Check out our latest news of the DyHealthNet project! It is a German-Italian collaboration where the team builds a network-based software platform for the analysis of population cohort data collected for a timespan of more than 10 years. 🩺🫀 www.mls.ls.tum.de/en/daisybio/...

2

DaiSyBio @daisybio.de · Dec 19

The whole DaiSyBio team wishes you happy holidays - more to come in 2025 ✨ (Nota bene: This is not an AI generated picture)

1 3

Reposted by DaiSyBio

Markus List @itisalist.bsky.social · Dec 14

Incredibly happy to finally see our manuscript "Emergence of power-law distributions in protein-protein interaction networks through study bias" published in @elife.bsky.social. doi.org/10.7554/eLif... It's been a long but fun journey with @dbblumenthal.bsky.social and @martinschaefer.bsky.social

Emergence of power-law distributions in protein-protein interaction networks through study bias

doi.org

1 9 16