Cathrine Sant (Petersen)
@cathrine_sant
Postdoc @GladstoneInst · Methods development for single-cell data analysis & applications to Alzheimer's
CHOIR is now online @NatureGenet! CHOIR is a new clustering method for single-cell data that evaluates whether clusters represent statistically distinct cell populations. CHOIR scales to millions of cells and works with single-/multi-omic data of any type! nature.com/articles/s4158…
CHOIR adds statistical muscle to single-cell #RNAsequencing, helping scientists at @GladstoneInst cluster cells more accurately and uncover real biological patterns—not just noise. rna-seqblog.com/choir-improves…
tSNE/UMAP preserve neighbors in a "soft" sense: Yes, tSNE recalls only ~30% of neighbors exactly. But ~60-80% of the 10 nearest tSNE neighbors are among the 100 nearest true neighbors! So even if some tSNE neighbors are not next-door neighbors, most are "from the same block".
No, tSNE/UMAP don't 'preserve neighborhoods'. Fig 2d shows that tSNE and UMAP get the large majority of k nearest neighbors wrong on all datasets. And this is 'recall' not even of true cell positions but just the logp1 mappings of the raw data!
Congratulations to @doctorcorces and the entire Gladstone team that hosted last week’s 2025 FunGen-AD meeting. The annual program examines genetic risk factors for Alzheimer’s disease, with the goal of finding new targets for preventing, diagnosing, and treating disease.
Gladstone scientists have developed a computational approach that accurately distinguishes cells of different identities or characteristics in complex biological samples used in diverse areas of biomedical and discovery research. @cathrine_sant @doctorcorces @NatureGenet
Gladstone scientists developed a computational approach that accurately distinguishes cells of different identities or characteristics in complex biological samples used in diverse areas of biomedical and discovery research. @NatureGenet @UCSF bit.ly/3FXBBZI
CHOIR is officially published @NatureGenet. No more arbitrary decisions in clustering. Add this to your single-cell toolbox - it will save you so much heart ache. And now it can scale to millions of cells. Compliments of the very talented @cathrine_sant.
CHOIR is now online @NatureGenet! CHOIR is a new clustering method for single-cell data that evaluates whether clusters represent statistically distinct cell populations. CHOIR scales to millions of cells and works with single-/multi-omic data of any type! nature.com/articles/s4158…
One of the last stories from my postdoc work with @HowardYChang and @WJGreenleaf is out today @ScienceMagazine led by Laksshman Sundaram. scATAC-seq of 8 different tumor types from TCGA. science.org/doi/10.1126/sc…
This paper adds to our studies anchored by solid neuropath data. Previously, we showed atypical AD = more regional tau burden related to symptoms. tinyurl.com/3yb6fy3e We conclude that selective vulnerability is key to deciphering atypical AD. See comments 👇 @cathrinepet
Why do some with #Alzheimer's pathology have more language or visual issues than memory problems? In a new paper led by @halsty & Renaud LaJoie, we explore if more copathology explains #AtypicalAlzheimer. tinyurl.com/mu8dtzts See comments 👇 @UCSFmac @GBHI_Fellows @AtypicalPIA
This study reveals something fascinating: Different packages (e.g., in terms of versions and programming languages) of the same algorithms/pipelines can lead to varied results in downstream analysis. Here are my thoughts: Firstly, this should not be used to critique Seurat and…
The choice of whether to use Seurat or Scanpy for single-cell RNA-seq analysis typically comes down to a preference of R vs. Python. But do they produce the same results? In biorxiv.org/content/10.110… w/ @Josephmrich et al. we take a close look. The results are 👀 1/🧵
Happening today!! Watch me try to describe my thesis work on clustering single-cell data in under 3 minutes! Live stream at graduate.ucsf.edu/watch-grad-sla…
New preprint! So excited to present CHOIR, a new clustering method for single-cell data that evaluates whether clusters represent statistically distinct cell populations. CHOIR works with both single- and multi-omic data of any type. Check it out! biorxiv.org/content/10.110…
The authors highlight that UMAP/tSNE preserve neighborhoods and conclude: "If used properly, 2D (tSNE/UMAP) embeddings will continue to help charting the world of single-cell biology." @hippopedoid @JanLause biorxiv.org/content/10.110…
Moving opening at #Tau2024 by Linde Jacobs on MAPT families fighting for a future. She and her sisters have already inspired work in my lab on MAPT mutation carriers and I hope her families' galvanizing story inspires more research curemaptftd.org/team/linde-jac…
In this preprint biorxiv.org/content/10.110… with @sindri_e we compared seven widely used methods for batch correction of single cell RNA-seq data. We found that all but one of the methods introduce batch effects when there are none. 1/N