Professor for Biomedical Network Science at Friedrich-Alexander University Erlangen-Nürnberg. [email protected], [email protected]
Erlangen, Deutschland
Joined November 2020
- Tweets82
- Following182
- Followers159
- Likes84
David B. Blumenthal retweeted
DataSAIL is out in @NatureComms
Since the preprint, we have improved the work a lot, thanks to countless reviewers and feedback.
You can find it here: nature.com/articles/s41467-0…
Thanks, @dbblumenthal and @ok55991, for helping and supervising me on this journey.
Interested in developing scalable data structures for biological data? Then apply for a position in the lab of our collaboration partner Johann Gamper in beautiful Bolzano. Deadline: Oct 30, noon. Details: dyhealthnet.ai/team/
Work with us and help us developing more reliable and bias-reduced protein-protein interaction networks in a great collaboration between Milan, Munich, and Erlangen
🚨 Postdoc or PhD opportunity! 🕸️ Protein-protein interaction networks face challenges like high error rates and biased degree of certain proteins. Help us improve their quality and develop computational methods to better interpret them! More info: schaeferlab.org/misc/join/"
David B. Blumenthal retweeted
In our latest @naturemethods perspective we provide 7 questions on how to detect data leakage in biological #MachineLearning applications. @judith_bernett @dbblumenthal @bit_tumcs @FloHasee @RomanJoeres @itisalist @TU_Muenchen #TUMCS #HSWT rdcu.be/dQuw1
David B. Blumenthal retweeted
Very proud to present our latest work: 7 guiding questions to avoid data leakage in biological machine learning applications ✨🔍 We hope that reflecting on these questions helps researchers to identify issues or shortcuts leading to overly optimistic performance estimates. 📈🧑🔬
A Perspective from @itisalist @judith_bernett @RomanJoeres @ok55991 @FloHasee @dg_grimm @bit_tumcs & @dbblumenthal discusses the issue of data leakage in machine learning models and presents 7 questions to identify and avoid problems as a result. nature.com/articles/s41592-0…
David B. Blumenthal retweeted
A Perspective from @itisalist @judith_bernett @RomanJoeres @ok55991 @FloHasee @dg_grimm @bit_tumcs & @dbblumenthal discusses the issue of data leakage in machine learning models and presents 7 questions to identify and avoid problems as a result. nature.com/articles/s41592-0…
David B. Blumenthal retweeted
So happy to announce that my paper with @itisalist and @dbblumenthal "Cracking the black box of deep sequence-based protein–protein interaction prediction" is finally published at Briefings in Bioinformatics doi.org/10.1093/bib/bbae076 ! So what is it about? 1/13 🧵
David B. Blumenthal retweeted
We are looking for a computational postdoc interested in cancer epigenomics to work with us in at the @IEOufficiale in Milan. Find more details on this and other positions here: schaeferlab.com/misc/join/. Please share
David B. Blumenthal retweeted
🧬Get ready for our #federated singular value decomposition (#SVD) algorithm, suitable for the #privacy and computational requirements of genome-wide association studies (#GWAS). A corresponding App is available in #FeatureCloud's #AppStoreHealthcare! 🧪 link.springer.com/article/10…
David B. Blumenthal retweeted
It's basically already old news but since Friday, DataSAIL is out as a preprint on @biorxivpreprint.
Now, you can split all your #biological #data for #ML without caring about Information or Data Leakage.
A summary in on the card below.
Check it out: doi.org/10.1101/2023.11.15.5…
David B. Blumenthal retweeted
Large-scale disease association databases are used in data-centric precision medicine but biased towards organ- & symptom-based disease definitions. @dbblumenthal @UniFAU aims to develop network-based approaches to disentangle data into new subsets, thereby improving predictions.
David B. Blumenthal retweeted
Ready to celebrate the next generation of scientists? Check out the €100k Early Career category of the #EinsteinFoundationAward2023 w/ @questbih! Get to know the brilliant finalists online on Nov 9 and find out who wins on Nov 14! @maxplanckpress (1/2)
David B. Blumenthal retweeted
Just a few more days until the deadline! ⌨️
How reproducible is @CVPR?
Let's find out together at #FAUhack on Dec. 8th-11th 2023 at @UniFAU
in Erlangen, Germany
with @anki_xyz and @VisionBernie
Register until October 31st!
reproducecvpr.tf.fau.eu/
David B. Blumenthal retweeted
Congratulations to my wonderful PhD student @KatochMitali for publishing today her first shared first-author paper in Acta Neuropathol Commun. 🥳💪🏻👏 #neuropath 🔬🧠 @IlaeWeb @yesILAE actaneurocomms.biomedcentral…
David B. Blumenthal retweeted
Fascinating keynote by Jingyi Jessica Li in her Overton prize lecture. Showing us that DE analysis after #clustering deserves some additional attention. #ISMBECCB2023
David B. Blumenthal retweeted
I am happy to announce that I am seeking for a #postdoctoral bioinformatician or computer scientist for a full, #permanent_position as a research fellow in my lab! Co-maintain popular #opensource projects like #Bioconda, #Snakemake, and #Varlociraptor!
koesterlab.github.io/#resear…
Check out our new #DrugstOne plug-in for online exploration of molecular networks and join our initiative: drugst.one/call/. Developing web frontends for systems medicine tools has never been easier!
#DrugstOne: Turn your #Bioinformatics tool into a powerful, interactive #DrugRrepurposing 💊 webtool in <15mins with 3 lines of code. Check out drugst.one/call/ and join our initiative and publication (arxiv.org/abs/2305.15453). @janbaumbach @MichiHartung @MajoreAndi
@aibe_fau goes digital history: “On the role of network topology in German-Jewish recommendation letter networks” out in @appliednetsci. Great collaboration with @erna_meyerink. 1/5
doi.org/10.1007/s41109-023-0…
(2) Can we characterize senders/addressees of (un)successful letters? Answers to both questions is yes. In particular, letters tended be more successful if sent by active recommenders and and addressing less sought after receivers. Full story in the paper. 4/5