Valentine Svensson

Read the full blog on our Substack: substack.com/home/post/p-19…

In our new blog post, we team up with @NVIDIAHealth to make it possible to map how drug effects cascades through gene regulatory networks.

It was a pleasure meeting members of the @BerkeleyML community last week! Our team always enjoys connecting with talented students who are passionate about ML x Bio 💙 We’re excited to continue these conversations and are open to speaking with prospective internship and full-time candidates. Check out our careers page (link in comments)

The one science reform we can all agree on, but we're too cowardly to do - experimental-history.com/p/the-one-scie…

Driftmachine: Spume & Recollection - open.spotify.com/album/5qFhCMq2…

A comprehensive analysis framework for evaluating commercial single-cell RNA sequencing technologies - doi.org/10.1093/nar/gk…

Probing the Feasibility of Single-Cell Fixed RNA Sequencing from FFPE Tissue - doi.org/10.3390/ijms27…

@arjunrajlab Should you correct for every test you've done in your life?

There is an interesting multiple hypothesis testing problem coming where you ask whether it's statistically valid to have a cool analysis if you run 100 (1000? 10,000?) analyses on a dataset and pick the best one.

1/ The technology behind @Tahoe_ai is now published in Nature Cancer. GENEVA: pool diverse disease models into one mosaic tumor → treat → deconvolve response at single-cell resolution. The architecture that had to exist before datasets like Tahoe-100M were possible.

After years of work, the centerpiece of my PhD is published in @NatureMethods! Read it to learn about the biophysical insights we can get from single-cell data! But first, I would like to talk a bit about RNA velocity and normalization. 1/

@Seyoon_L I wanted to find examples in a couple of papers, or a reference discussing it, but couldn't. The kind of thing you see when analyzing, but don't really see if looking at formatted results.

@vallens Thank you for this! Been troubled by low count cells for years. It’s always that radial thing, low in the middle and higher out toward the edges, even across clusters. I’ve been filtering them out bits by bits, but the pattern just kept coming back. Would love to try this!

Improving SCVI for low-count cells through self-supervised augmentation - nxn.se/p/improving-sc…

@BiomedErs Awesome! Let me know when you have the page so I can subscribe to the RSS

@vallens Love your blog. Thank you for writing these. Inspired me to start a blog as well, writing my first post now. As a first year PhD I do not have your experience, but will get there haha 💪

Interferons in health and disease - cell.com/cell/abstract/…

The Misuses of the University - publicbooks.org/the-misuses-of…

The Origins of Adjuvants - press.asimov.com/articles/adjuv…

A hyperparameter benchmark of VAE-based methods for scRNA-seq batch integration biorxiv.org/content/10.648…

Improving SCVI for low-count cells through self-supervised augmentation biorxiv.org/content/10.648… #biorxiv_bioinfo

What if we scored drugs not by whether they kill diseased cells, but by whether they push them back to normal? New blog post from Tahoe: we applied cell-state reversal to colorectal cancer using Tahoe-100M, and the results are striking.

The ‘Gravity Falls’ series finale turns 10 years old this month.