Author

Niklas Köhler

Helmholtz Munich

Listed by name: these papers carry no ORCID for this author, so different people with the same name may be mixed together.

Biology Disciplines

Click to filter author papers

Bioinformatics· 3 Systems Biology· 1

Signature Assays, Organisms & Methods

Click to filter · or open tag

Organism:human· 4 Research Focus:cell type annotation· 2 Assay:imaging· 2 Research Focus:benchmarking· 1 Approach:contrastive learning· 1 Approach:deep learning· 1 Approach:experimental assay· 1 Approach:machine learning (classical)· 1

Frequent Co-Authors

Fabian Joachim TheisiD· 4 shared Andreas‐David BrunneriD· 1 shared Andrew J. FilbyiD· 1 shared Anne Elizabeth CarpenteriD· 1 shared Ben AsaniiD· 1 shared Eugenia VoytikiD· 1 shared F. Alexander WolfiD· 1 shared Florian MeieriD· 1 shared

Activity by Year

2017-2021 · peak 2/yr

Co-Author Collaboration Network

10 top collaborators · 19 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

4 shared papersBioinformatics

Fabian Joachim TheisORCID

Mutual co-authors in cluster

Author profile →

4 papers

Recent·Most cited

Deep learning does not outperform classical machine learning for cell-type annotation

Niklas Köhler, Maren Büttner, Niry Andriamanga, Fabian Joachim Theis

This study evaluates 10 classification methods across 14 datasets and finds that deep learning does not outperform classical machine learning for cell-type annotation.

bioRxiv · Jun 24, 2021 · PreprintBioinformatics

Deep learning the collisional cross sections of the peptide universe from a million training samples

Florian Meier, Niklas Köhler, Andreas‐David Brunner, Jean-Marc H. Wanka, Eugenia Voytik, Maximilian T. Strauss, Fabian Joachim Theis, Matthias Mann

The paper measures CCS values for a million peptides across five organisms and develops a method to predict CCS for any peptide and organism.

Nature Communications · Feb 19, 2021Peer-reviewedSystems Biology

Self-supervised retinal thickness prediction enables deep learning from unlabeled data to boost classification of diabetic retinopathy

Olle Holmberg, Niklas Köhler, Thiago Gonçalves dos Santos Martins, Jakob Siedlecki, Tina Herold, Leonie Franziska Keidel, Ben Asani, Johannes Benedikt Schiefelbein, Siegfried Georg Priglinger, Karsten Kortuem, Fabian Joachim Theis

The paper proposes a self-supervised method to predict retinal thickness from fundus images, enabling efficient diabetic retinopathy classification with fewer annotations.

Nature Machine Intelligence · Nov 9, 2020Peer-reviewedBioinformatics

Reconstructing cell cycle and disease progression using deep learning

Philipp Eulenberg, Niklas Köhler, Thomas Blasi, Andrew J. Filby, Anne Elizabeth Carpenter, Paul C. T. Rees, Fabian Joachim Theis, F. Alexander Wolf

This paper uses deep learning to reconstruct cell cycle stages and disease progression from image data, achieving lower error rates and faster analysis.

Nature Communications · Aug 31, 2017Peer-reviewedBioinformatics