Dmitry Kobak

dmitry kobak photo

dmitry.kobak@uni-tuebingen.de
Google Scholar
@amoeba at CrossValidated.SE

Current main interests: exploratory analysis of single cell RNA-seq data, dimensionality reduction, clustering. t-SNE!


Nano CV

From 2017: Postdoc in Berens lab at Tübingen University.
Current work: biorxiv 2018a, biorxiv 2018b

2013–2016: Postdoc in Machens lab at Champalimaud Institute, Lisbon.
Main paper: eLife 2016.

2007–2012: PhD in Mehring lab at Freiburg University, then at Imperial College London.
Main paper: Jneurosci 2012.

Before that studied computer science (BSc) and theoretical physics (MSc) in St. Petersburg.


paper pdfs

How do children manage to pull their parents together in a PCA projection of a GWAS data set?

15 November 2018

Family tetrahedron

See https://stats.stackexchange.com/questions/376381 for details.


The art of using t-SNE for single-cell transcriptomics

01 November 2018

Putin's peaks

See our preprint The art of using t-SNE for single-cell transcriptomics for details. See also https://github.com/berenslab/rna-seq-tsne/.


Putin's peaks

25 April 2018

Putin's peaks

See our preprint Putin’s peaks: Russian election data revisited for details. See also https://github.com/dkobak/elections/.

Update (May): this paper has been published in Significance. This is a follow-up to our 2016 Significance paper Statistical fingerprints of electoral fraud? which itself was a popular companion paper to a more formal 2016 AOAS paper Integer percentages as electoral falsification fingerprints.


Unit-variance ridge regression

17 April 2018

Unit-variance ridge regression

See https://stats.stackexchange.com/questions/340099 for details.


Sparse reduced-rank regression for patch-seq

16 April 2018

Sparse reduced-rank regression for exploratory visualization of single cell patch-seq recordings

See our preprint Sparse reduced-rank regression for exploratory visualization of single cell patch-seq recordings for details.