Joon Sang Lee (@JSLee198) profile picture

Cellular Analysis Group Leader at Sanofi Precision Oncology

Cambridge
Joined February 2023
Joon Sang Lee retweeted
How well do single-cell foundation models perform w/o finetuning? Our work @GenomeBiology shows that in zero-shot settings, scGPT and Geneformer often underperform traditional methods, raising questions about their utility for biological discovery. 📰 genomebiology.biomedcentral.…
6
48
2
247
25,900
Joon Sang Lee retweeted
Beautiful research with important translational implications by Brian Brown and @MiriamMerad 's groups showing how tumor-secreted IL4 leads to immune evasion through tumor macrophages cell.com/cell/abstract/S0092…
1
29
1
109
9,027
Joon Sang Lee retweeted
Online Now: Cross-tissue human fibroblast atlas reveals myofibroblast subtypes with distinct roles in immune modulation dlvr.it/TDTSGq
1
64
204
17,401
Joon Sang Lee retweeted
Thrilled to share our „PRedictor Of PHEnoTypes“ model Prophet! Led by @Alejandro__TL & @_yji_, Prophet is a transformer-based model that predicts outcomes for unseen experiments. It aims to understands biology by learning across assays and phenotypes over 4.7M+ experiments.
1
64
4
220
31,465
Joon Sang Lee retweeted
In this blog post, I use our recent SEraster #Rstats tool to perform spatial bootstrapping to evaluate the stability of cell-type proportion estimates from #singlecell #spatialtranscriptomics data. Try it out for yourself: jef.works/blog/2024/07/23/sp… #bioinformatics #codetutorial
27
89
9,117
Joon Sang Lee retweeted
Check out our just published SEraster tool for aggregating #SingleCell #SpatialTranscriptomics data into square/hex pixels for scalable spatial #DataAnalysis. Paper: academic.oup.com/bioinformat… Software: github.com/JEFworks-Lab/SEra… #RStats Congrats to @Gohta_Aihara + team 🥂 See 🧵👇
Having trouble analyzing large-scale #spatialomics data? SEraster, a rasterization preprocessing framework on #Rstats that aggregates cellular information into spatial pixels for scalable analysis, is now published on Bioinformatics! Paper: doi.org/10.1093/bioinformati… 🧵👇 1/n
13
57
8,514
Joon Sang Lee retweeted
REVIEW @NatImmunol @CaitlinZebley T cell dysfunction and therapeutic intervention in cancer nature.com/articles/s41590-0…
78
255
26,042
Joon Sang Lee retweeted
scType Marker-based cell type annotation for #SpatialTranscriptomics Visium Slide-seq Use positive + negative markers (from scTypeDB database with previously scRNAseq-identified cell type markers or custom marker genes), instead of deconvolution. R & Python github.com/kris-nader/sp-typ… or github.com/kris-nader/sc-typ… Mitro Miihkinen lab Bioinformatics 2024 @mitroe academic.oup.com/bioinformat…
39
157
18,713
Joon Sang Lee retweeted
Immune cell triads reprogram exhausted CD8+ T cells for effective tumor elimination dlvr.it/T9dfhK
2
86
2
320
39,594
Joon Sang Lee retweeted
Cytotoxic T cells form topographically complex, undulatory immune synapses with their targets to promote killing @SciImmunology @huselab science.org/doi/10.1126/scii…
1
20
81
10,689
Joon Sang Lee retweeted
IL-12/15/18 differentially activate NK cell subsets driving unique memory-like programs with implications for NK cell therapy @SciImmunology Congrats @FoltzPhd @LabFehniger @Jenn__Tran science.org/doi/10.1126/scii…
2
32
1
133
12,645
Joon Sang Lee retweeted
Type I conventional DCs facilitate immunotherapy in pancreatic cancer @ScienceMagazine science.org/doi/10.1126/scie…
13
49
2,930
Joon Sang Lee retweeted
Replying to @ScholarshipfPhd
The paper discusses the use of AI language bots, such as ChatGPT, in writing scientific manuscripts. It highlights both the potential benefits and concerns associated with using these tools in the scientific publishing process. openread.academy/en/paper/re…
4
16
1,971
Joon Sang Lee retweeted
🌟 Join me at #Scverse2024 for the "GPU-Accelerated Single-Cell Analysis with rapids-singlecell" workshop! 📍 Munich, Germany 🗓️ Sept 10-12 Don't miss out! Register now: [scverse.org/conference2024/] Explore the toolkit: [github.com/scverse/rapids_si…] @scverse_team
2
11
42
3,845
What would you do with 1000+ spatial transcriptomics samples with corresponding H&E-stained whole-slide images? Meet HEST-1k, a collection of 1,108 ST samples assembled from 131 public and internal cohorts encompassing 25 organs, 2 species. HEST-1k includes over 1.5 million expression–morphology pairs. 🔍Explore 3 use-cases for HEST-1k: - HEST-Benchmark: Evaluate gene expression prediction from histology across 10 organs and 9 cancer types, testing multiple foundation models for pathology including UNI, and GigaPath. - HEST for discovery: Explore our proof-of-concept for multimodal biomarker characterization using Xenium breast cancer samples. - HEST for fine-tuning pathology foundation models: See how HEST-1k can enhance foundation models for histology with expression-guided fine-tuning. 📄Preprint: arxiv.org/pdf/2406.16192 👩‍💻 Code and Data access: github.com/mahmoodlab/HEST Congratulations to @guillaumejaume, @pauldoucet, and everyone else who contributed to this work. Huge thanks to everyone who helped curate the dataset. #SpatialTranscriptomics #ComputationalPathology #CancerResearch #Bioinformatics
3
104
5
378
50,706