nxn

TSNE is a method for visualizing high-dimensional data that preserves local neighborhoods on a two-dimensional plane.

I was thinking about my reading habits.

There are no actual rules about the order of authorships of papers in academic journals.

Occasionally, when a paper is published with a new interesting dataset, the GEO accession is locked as ‘private’ until a certain release date.

I was reminded that about a year ago (May 2025) I investigated whether LLM-based website crawlers could be used to extract how many cells were collected for individual single-cell papers.

The 10x Genomics Flex kit for single-cell RNA-sequencing relies on a fixed set of gene probes as opposed to untargeted RNA-sequencing.

Last year, the company I work for rebranded as Tahoe Therapeutics.

To improve utility of cells with low UMI counts, I wanted to optimize KL weighting in SCVI models.

TSNE visualization is useful to explore data but is time consuming without access to an Nvidia GPU.

The single cell analysis package scvi-tools allow you to model data quickly and easily if you have access to a GPU.

In this post I motivate why I think it would be useful to have clear thresholds for when a gene is not expressed, test a strategy to obtain such thresholds, and learn it doesn’t work.

Anthropic recently launched Claude for Life Sciences, which in practical terms means a marketplace of plugins, MCP’s, and agent skills focused on tasks you encounter in biomedical research.

With a trained SCVI model you can perform differential expression analysis to learn which genes are increased and decreased in gene expression as cells change their states.

Time is special.

I wanted to speed up negative binomial regression with a novel transformer model, but it actually turns out that a classical method of moments is a better solution.

I’m working on a project where I want to use a lot of well-annotated scRNA-seq data.

Generally, neural networks are nonlinear function approximators.

In the game Blue Prince , every run of the game have you fill a mansion with different rooms to wander through.

For last couple of months I have been exploring batch integration strategies with SCVI and MRVI, and the possibility to optionally disable integration when encoding single cells.

For the past ~7 years I have been tracking the scale of scRNA-seq experiments in a spreadsheet which we published as the ‘Single Cell Studies Database’ a few years ago (Svensson, da Veiga Beltrame, and Pachter 2020).