Author

Arkadij Kummer

ETH ZurichORCID 0000-0002-1991-8036Unclaimed ORCID profile · follow to get notified of new preprints

Biology Disciplines

Click to filter author papers

Bioinformatics· 2 Molecular Biology· 1

Signature Assays, Organisms & Methods

Click to filter · or open tag

Research Focus:benchmarking· 2 Assay:protein sequence· 2 Organism:bacteria (other)· 1 Approach:Bayesian inference· 1 Approach:deep learning· 1 Research Focus:differential expression· 1 Research Focus:drug response / drug discovery· 1 Approach:machine learning (classical)· 1

Frequent Co-Authors

Eric J. MaiD· 3 shared Edison OngiD· 1 shared Farbod MahmoudinobariD· 1 shared Gwo‐Yu ChuangiD· 1 shared Hao ZhouiD· 1 shared Isha AroraiD· 1 shared J. Wade DavisiD· 1 shared Mihaela G. GadjevaiD· 1 shared

Activity by Year

2021-2026 · peak 1/yr

Co-Author Collaboration Network

10 top collaborators · 25 mutual co-author links

Click any collaborator node to inspect or filter shared papers

Focal author (center)Selected co-authorORCID co-authorUnlinked name

Curved chords = mutual co-authorship

3 shared papersBioinformatics

Eric J. MaORCID

Mutual co-authors in cluster

Author profile →

3 papers

Recent·Most cited

BioinformaticsbioRxivPreprintSep 16, 2026

A Turing-Style Test for In-Silico Antibodies: How Sampling Mode Makes WGAN-GP Beat VAE in the Wet Lab

Arkadij Kummer, Farbod Mahmoudinobar, Wendi Liu…·Sandeep Kumar

This paper compares GANs and VAEs for generating viable antibody sequences, finding GANs outperform VAEs when sampling strategies are optimized.

DOI 10.64898/2026.09.11.750936

protein sequencedeep learning

𝕏 PostBluesky

BioinformaticsComputers in Biology and MedicinePeer-reviewedFeb 10, 2024

mtx-COBRA: Subcellular localization prediction for bacterial proteins

Isha Arora, Arkadij Kummer, Hao Zhou…·Edison Ong

mtx-COBRA improves subcellular localization prediction for bacterial proteins, outperforming PSORTb in accuracy for proteins with unknown localization.

8 citations

bacteria (other)protein sequence

𝕏 PostBluesky

Molecular BiologyEntropyPeer-reviewedJun 8, 2021

Principled decision-making workflow with hierarchical Bayesian models of high throughput dose-response measurements

Eric J. Ma, Arkadij Kummer

The paper applies hierarchical Bayesian models to high throughput protein melting point data to enable principled decision-making and compares it to maximum-likelihood curve fitting.

3 citations

protein structureBayesian inference

𝕏 PostBluesky