We also applied Modanovo to a dataset of human cells infected with monkeypox virus, where it complemented database-driven approaches and revealed biologically relevant phosphosites, highlighting its utility in real-world discovery.

Modanovo achieves strong performance, successfully sequencing peptides carrying biologically relevant PTMs.

Modanovo builds on Casanovo, a transformer model for peptide sequencing. We fine-tuned ot end-to-end expanding Casanovo’s token vocabulary and corresponding layers.

Why Modanovo?
Existing tools often struggle with PTMs, yet PTMs are central to protein regulation and disease. We built a model for 19 amino acid-PTM combinations, leveraging data from in vivo experiments and from PROSPECT-PTM.

Update of our protein outlier caller PROTRIDER. We now handle missing values, a widespread issue for mass spec where missing values are not a random -- and this improves outlier detection on non-missing data! Thumbs up to Daniela and George for the great work.
doi.org/10.1101/2025...

and much more... poster 128

In today's poster session #probgen25. To the pop gen folks, interesting observation: The influence of a nucleotide on reconstructing others, rather than its own reconstructability, is a better predictor of function. This metric makes DNA LMs beat conservation in several benchmarks.

and @pedrotomazdasilva.bsky.social will present tomorrow at #probgen25 poster 128 on dependency analysis of DNA language models. Come and see what functional relationships DNA LMs capture, from regulatory code to RNA structures. Preprint: doi.org/10.1101/2024...

Tomorrow Johannes Hingerl @johahi.bsky.social gives a talk on scooby at #probgen25. Enjoy learning in the legendary CSHL auditorium how to model RNA-seq and ATAC-seq profiles in individual cells from half a megabase of genomic sequence. Preprint:
doi.org/10.1101/2024...

Hello #probgen25! We have 3 contribs this year @lauradmartens.bsky.social starts today, poster 87, presenting scooby modeling scRNA-seq and sc-ATAC-seq profiles from DNA and applications. Shhh... don't tell it further... rumour says there are awesome cute scooby stickers to win ;-)

Benchmarking with orthogonal data matters: PROTRIDER outperforms alternative approaches at detecting these rare variants affecting protein abundance

Protein outliers are enriched in variants predicted to be pathogenic by variants likely triggering NMD and by AlphaMissense.

We applied PROTRIDER to a mitochondrial disorder dataset and two tumor cell line panels. Here is an example of disease-causing protein in an individual affected by a mitochondrial disorder.

To assess the statistical significance of the residuals, we compute P-values from the Student’s t-distribution which yielded much better statistical calibration than the normal distribution.

PROTRIDER uses an autoencoder to capture known and unknown sources of protein intensity variations yielding expected log-intensities for each protein in each sample. Deviations of the measurements to these expected values are used to call outliers.

On Wed., Pedro @pedrotomazdasilva.bsky.social talks about interpreting DNA Language models through nucleotide dependency analyses. Capturing functional interactions -maps like this one for an intron- and more: RNA structures, reg motifs,... preprint:
www.biorxiv.org/content/10.1...