Curious? Try it yourself and leave a star ⭐️ on GitHub
github.com/usiGrabber/u...
🧵 7/7

In under 2 days, we construct a PTM-specific dataset of nearly 11 million spectra and used it to retrain a phosphorylation classifier. The retrained model matched the performance of the original model on an independent test set, showing the power of scalable, automated data generation.
🧵 6/7

Within 49 hours we parsed over 800 million PSMs from over 1,200 projects on PRIDE and made them filterable through their metadata, allowing the curation of task-specific datasets.
🧵 5/7

usiGrabber 🏗️ to the rescue! A scalable framework for assembling large proteomic datasets. It extracts spectra identification, stores project-level metadata, indexes raw spectra using USIs, and offers download utilities to retrieve spectra data at scale.
🧵 4/7

Curating a task-specific proteomics dataset requires deep domain expertise. In a time consuming step, researchers have to manually select relevant spectra from accessible repositories, potentially missing out important data and projects.
🧵 3/7

G. Auge, M. Clausen, K. Ketterer, J. Schaefer, N. Schmitt, T. Altenburg, @yhartmaring.bsky.social, @hendraet.bsky.social, C.N. Schlaffner & B.Y. Renard turned what started as a fun hackathon 💻 into this project, huge thanks to everyone who brainstormed, coded, and experimented along the way!🙌
🧵 2/7

We are excited to present 🏗️ 'usiGrabber: Automating the curation of proteomics spectra data at scale, making large datasets ready for use in machine learning systems' now available on bioRxiv:
doi.org/10.64898/202...
#proteomics #machine-learning #mass-spec #dataset-curation

🧵 1/7

Despite hundreds of published models for cancer drug response prediction, none are used in clinical practice.

Why?

In our preprint, we outline key challenges and introduce a living benchmark to help the field move forward.

Check out the thread by my co–first author @judith-bernett.bsky.social:

@y.hartmar

What a run! 🏃‍♀️🏃‍♂️
Last week, our group took part in the Berlin #Firmenlauf, and we had an amazing time out there, running, cheering each other on, and just enjoying the energy of the event! #TeamVibes

Thanks to everyone who made it such a memorable day 🤗 #ScienceInSneakers @hpi.bsky.social

We’re excited to share that our paper on rapid molecular classification of brain tumors has just been published in Nature Medicine!

In his project in collaboration with the Icahn School of Medicine at Mount Sinai @hpims.bsky.social, he is exploring the effect of biological administration timing on treatment outcomes in patients with Inflammatory Bowel Disease. Supervised by Dr. Susanne Ibing @sibing.bsky.social.

Akin joyously standing in front of his poster titled “Morning administration of biological therapy is associated with decreased risk of treatment failure in patients with Inflammatory Bowel Disease” at the European Crohn’s and Colitis Organisation (ECCO) congress!

We're proud to have Akin present his ongoing master's thesis project at the European Crohn's and Colitis Organisation (ECCO)
Congress 2025!

Check out the awesome opportunity to join the group of one of the brilliant minds at our chair!

Her work produces daily immunity indices for each regional location, given vaccination and infection data for different infectious diseases, which in turn can be used to, e.g., predict the spread of disease 🧬🤧🌐

Ferdous Nasri presenting her work on the 'Spatio-temporal immunity index tool for infectious diseases' at the Digital Health Summit at University of Cape Town in South Africa!

She has published numerous papers and presented her work at various conferences.

Her next step will focus on her research collaborations across the atlantic. 🚀

Her achievements are outstanding and we are very proud to be celebrating this day with her!💐

She started her PhD as a part of the Böttinger Lab, spent some time working with Mount Sinai in NY and switched to our Lab later. She led projects with many students, one of which won the DMEA Sparks Award🏆

Dr. Susanne Ibing joyously shaking hands with Prof. Bernhard Renard!

Susanne defends her dissertation in front of the fully-packed room.

Dr. Susanne standing proudly wearing her graduation hat and robe in front of the HPI Digital Health building smiling with all the people that participated in her dissertation defence.

💐Congratulations to ✨ Dr. Susanne Ibing! 🎓

Susanne, @sibing.bsky.social, successfully defended her doctoral dissertation @hpi.bsky.social this month on “Computational Strategies for Chronic Disease Characterization and Treatment by Leveraging Electronic Health Records and Omics Data” 🩻🧬👩🏼‍💻

With an ablation study, we demonstrated the added value of information derived from clinical notes not only for the computable phenotyping, but also the disease prediction task.

🧵 6/6

When comparing coded conditions between identified cases and controls, we saw significant overrepresentation of GI-related conditions in cases, indicating the diagnostic delay of the disease.

🧵 5/6

We found that adding information on age at diagnosis extracted from the clinical notes improves the phenotyping performance and allows to better distinguish between referral and incident cases, compared phenotyping mainly relying on structured clinical data.

🧵 4/6

For automated cohort identification, we compared two computable phenotyping approaches with different levels of NLP and information extracted from clinical notes incorporated.

🧵 3/6

Diagnostic delay is a common problem in Crohn’s disease, and with delayed treatment induction leading to overall worsened outcomes. This study aimed to automatically identify newly diagnosed patients to use pre-diagnostic EHR for data disease prediction.

🧵 2/6

New paper: Electronic Health Records-Based Identification of Newly Diagnosed Crohn’s Disease Cases

By Susanne Ibing @sibing.bsky.social , Julian Hugo, et al., now available in Artificial Intelligence in Medicine.

authors.elsevier.com/a/1kA113KEGa...

🧵 1/6 Thread below:

Picture of the group members smiling standing in a meeting room facing the camera.

Happy New Year and a warm hello world from our group! 🥳😄