Tessera

Judith Bernett

@judith-bernett.bsky.social

Bioinformatics PhD student at TUM 🧬🖥️ 👩‍💻 Currently interested in machine learning pitfalls, drug response prediction, and protein-protein interaction prediction.

66 Posts106 Follower 103 Folgt

Judith Bernett @judith-bernett.bsky.social · 2mo ago

New release (small bugfix!)

Psyduck Relaxing in Floaty

ALT: Psyduck Relaxing in Floaty

static.klipy.com

nf-core@nf-co.re · 2mo ago

Pipeline release! nf-core/drugresponseeval v1.2.2 - Drugresponseeval 1.2.2 - Pretty Psyduck! Please see the changelog: https://github.com/nf-core/drugresponseeval/releases/tag/1.2.2

Judith Bernett @judith-bernett.bsky.social · 3mo ago

New pipeline version out now! Also, our paper has finally been published in Nature Communications, come check it out! www.nature.com/articles/s41...

Snorlax Sleeping in the Woods

ALT: Snorlax Sleeping in the Woods

static.klipy.com

nf-core@nf-co.re · 3mo ago

Pipeline release! nf-core/drugresponseeval v1.2.1 - Drugresponseeval 1.2.1 - Sleepy Snorlax! Please see the changelog: https://github.com/nf-core/drugresponseeval/releases/tag/1.2.1

Judith Bernett @judith-bernett.bsky.social · 8mo ago

New pipeline version out now!!

a cartoon character is standing on a dirt path

ALT: a cartoon character is standing on a dirt path

media.tenor.com

nf-core@nf-co.re · 8mo ago

Pipeline release! nf-core/drugresponseeval v1.2.0 - Drugresponseeval 1.2.0 - Gentle Togepi ! Please see the changelog: https://github.com/nf-core/drugresponseeval/releases/tag/1.2.0

Judith Bernett @judith-bernett.bsky.social · last yr.

New pipeline version finally out now! 🥳

a cartoon of a bird with a lightning bolt coming out of it 's wings

ALT: a cartoon of a bird with a lightning bolt coming out of it 's wings

media.tenor.com

nf-core@nf-co.re · last yr.

Pipeline release! nf-core/drugresponseeval v1.1.0 - Drugresponseeval 1.1.0 - Humongous Zapdos! Please see the changelog: https://github.com/nf-core/drugresponseeval/releases/tag/1.1.0

Judith Bernett @judith-bernett.bsky.social · last yr.

🧬🖥️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.

Judith Bernett @judith-bernett.bsky.social · 2y ago

@lisiarend.bsky.social, @quirinmanz.bsky.social, Kilian Kirmes, and I just created a nice overview of a typical CyTOF experiment using BioRender - now also available there as a template - and on Wikipedia! 🎨✨ Check it out and feel free to use it! en.wikipedia.org/wiki/CyTOF

Judith Bernett @judith-bernett.bsky.social · 2y ago

🧬🖥️ 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.

Judith Bernett @judith-bernett.bsky.social · 2y ago

🧬🖥️Tranferring my papers, Pt. 2: "Guiding questions to avoid data leakage in biological machine learning applications" doi.org/10.1038/s41592-024-02362-y, which I published with @itisalist.bsky.social @dbblumenthal.bsky.social @romanjoeres.bsky.social, D. Grimm, F. Haselbeck, and O.V. Kalinina. 🧵1/20

Overview figure of the seven guiding questions addressed in manuscript regarding (1) Study bias, (2) Label distribution, (3) Data shift, (4) Inter-sample similarities, (5) Unavailable features, (6) Features as illegitimate surrogates, (7) Pretraining

Judith Bernett @judith-bernett.bsky.social · 2y ago

Update: Since we published "Cracking the black box of deep sequence-based protein-protein interaction prediction," we have received mostly positive feedback. I presented our work at multiple conferences, including the ISMB/ECCB23 and GCB23. 🧬🖥️ doi.org/10.1093/bib/...

Cracking the black box of deep sequence-based protein–protein interaction prediction

Abstract. Identifying protein–protein interactions (PPIs) is crucial for deciphering biological pathways. Numerous prediction methods have been developed a

academic.oup.com

Judith Bernett @judith-bernett.bsky.social · 2y ago

🧬💻Transferring my paper posts here, starting off with "Cracking the black box of deep sequence-based protein-protein interaction prediction" doi.org/10.1093/bib/..., which I published together with @itisalist.bsky.social and @dbblumenthal.bsky.social. So what was it about? 1/13 🧵

Overview figure summarizing the major analyses of the paper. In the setting employed by published methods, performances are strongly inflated regardless of model and dataset. When the positive training dataset is randomly rewired, there is only a slight decrease in performance. When the models are evaluated on unseen proteins that are unsimilar to the training proteins, performances become random.

Markus List @itisalist.bsky.social · 2y ago

Had great fun attending the #gcb2024 in Bielefeld. Our group had the honor of contributing 2 talks (Judith Bernett on guiding questions to avoid data leakage and Johannes Kersting on GRN inference with DysRegNet. A surprise at the end: Alex Dietrich won the poster prize with the project omnideconv!