Lorin Crawford
@lcrawford.bsky.social
📤 903
📥 284
📝 13
Principal Researcher in BioML at Microsoft Research.
pinned post!
In our newest preprint, we show that simply increasing the size of pre-training datasets doesn't necessarily improve the performance of single-cell foundation models on downstream tasks. Really proud of
@alandenadel.bsky.social
for leading this effort! See his thread below for more details👇
add a skeleton here at some point
about 1 year ago
1
5
2
Excited to see our hierarchical cross-entropy strategy for improved cell type annotation out today in
@natcomputsci.nature.com
! Congratulations to the team! Try out the HCE loss with your own models and datasets: 📄:
www.nature.com/articles/s43...
💻:
github.com/microsoft/hc...
loading . . .
Improving atlas-scale single-cell annotation models with hierarchical cross-entropy loss - Nature Computational Science
A hierarchical cross-entropy loss is presented, which incorporates ontology structure into training and improves the out-of-distribution performance of large-scale single-cell annotation models withou...
https://www.nature.com/articles/s43588-025-00945-z
28 days ago
0
4
1
reposted by
Lorin Crawford
Nature Computational Science
28 days ago
📢Out now! Sebastiano Cultrera di Montesano, Peter S. Winter,
@lcrawford.bsky.social
, and colleagues present a hierarchical cross-entropy loss that improves performance of single-cell annotation models.
www.nature.com/articles/s43...
🖥️ 🧬
loading . . .
Improving atlas-scale single-cell annotation models with hierarchical cross-entropy loss - Nature Computational Science
A hierarchical cross-entropy loss is presented, which incorporates ontology structure into training and improves the out-of-distribution performance of large-scale single-cell annotation models withou...
https://www.nature.com/articles/s43588-025-00945-z?utm_source=bluesky&utm_medium=social&utm_campaign=natcomputsci
0
2
2
Come do a research internship with us! Details on the posting and how to upload a research statement given below!
add a skeleton here at some point
3 months ago
0
0
1
reposted by
Lorin Crawford
C. Brandon Ogbunu
3 months ago
"What James Watson got wrong about DNA" By the great Sohini Ramachandran (
@sramach.bsky.social
) and your boy for The Boston Globe (
@bostonglobe.com
).
www.bostonglobe.com/2025/11/14/o...
loading . . .
What James Watson got wrong about DNA - The Boston Globe
The science he helped pioneer consistently undermines his view that genes determine everything about us.
https://www.bostonglobe.com/2025/11/14/opinion/james-watson-dna-research/?event=event12
5
136
72
Alan did a ton of work on our latest revision. Message of the story still holds: more data does not necessarily lead to better downstream performance for many models. Check out all our new analyses here:
www.biorxiv.org/content/10.1...
add a skeleton here at some point
4 months ago
0
2
1
Come do an internship with us! Please apply
add a skeleton here at some point
4 months ago
1
4
2
reposted by
Lorin Crawford
arXiv q-bio.GN Genomics
7 months ago
Gindra, Palla, Nguyen, Wagner, Tran, Theis, Saur, Crawford, Peng: A Large-Scale Benchmark of Cross-Modal Learning for Histology and Gene Expression in Spatial Transcriptomics
https://arxiv.org/abs/2508.01490
https://arxiv.org/pdf/2508.01490
https://arxiv.org/html/2508.01490
0
0
6
reposted by
Lorin Crawford
The American Journal of Human Genetics
7 months ago
📣Online now! 📄Sparse modeling of interactions enables fast detection of genome-wide epistasis in biobank-scale studies 🧑🤝🧑
@lcrawford.bsky.social
@julian-stamp.bsky.social
& co
loading . . .
Sparse modeling of interactions enables fast detection of genome-wide epistasis in biobank-scale studies
The sparse marginal epistasis test overcomes computational limitations of previous mapping approaches by focusing its epistatic search to regions of the genome that have some known functional relation...
https://www.cell.com/ajhg/fulltext/S0002-9297(25)00274-5
1
6
2
reposted by
Lorin Crawford
Kevin K. Yang 楊凱筌
9 months ago
Do protein language models store different structural elements in factorizable subnetworks? To find out, we masked out PLM weights to suppress performance on CATH subcategories or secondary structure elements while maintaining performance on other sequences or residues.
1
26
5
reposted by
Lorin Crawford
Amstat-American Statistical Association
10 months ago
Members have elected Brian Millen as the 122nd president of the association. Also elected: Julia Sharp, Vice President; Martin Slawski, Council of Sections Representative; Ruixiao Lu, Council of Chapters Representative; Pedro Silva, International Representative. Congratulations!
tinyurl.com/ptunrn5z
0
3
1
reposted by
Lorin Crawford
National Institute of Statistical Sciences
11 months ago
"Leadership in Trustworthy AI" COPSS-NISS Leadership Webinar with David Donoho (Stanford), Michael I. Jordan (UC Berkeley), and Tracy Ke (Harvard) - online Tuesday, April 29, 2025 at 12-1pm ET. Register at:
www.niss.org/events/copss...
0
2
3
Great to see our paper presenting recall, a framework which calibrates clustering for the impact of data "double-dipping" in single-cell studies, out today in AJHG! Congratulations,
@alandenadel.bsky.social
and co-authors!
add a skeleton here at some point
12 months ago
1
9
3
reposted by
Lorin Crawford
Nature Reviews Genetics
12 months ago
New online! Adapting systems biology to address the complexity of human disease in the single-cell era
loading . . .
Adapting systems biology to address the complexity of human disease in the single-cell era
Nature Reviews Genetics, Published online: 10 March 2025; doi:10.1038/s41576-025-00821-6Differences between humans and experimental models create a translational gap that makes it difficult to extrapolate research findings. The authors review…
https://www.nature.com/articles/s41576-025-00821-6?utm_source=dlvr.it&utm_medium=bluesky
0
8
3
reposted by
Lorin Crawford
about 1 year ago
🔥 Benchmark Alert! MotifBench sets a new standard for evaluating protein design methods in motif scaffolding. Why does this matter? Reproducibility & fair comparison have been lacking—until now. Paper:
arxiv.org/abs/2502.12479
| Repo:
github.com/blt2114/Moti...
A thread ⬇️
1
41
22
reposted by
Lorin Crawford
RECOMB Conference Series
about 1 year ago
🎉Congratulations to the authors!🎉 The list of accepted papers for
#RECOMB2025
is now live ➡️
recomb.org/recomb2025/a...
What are your plans for Seoul? Let us know!
www.youtube.com/watch?v=uztj...
#seoul
loading . . .
RECOMB 2025 | PROGRAM
RECOMB 2025 - Yonsei University, Seoul
https://recomb.org/recomb2025/accepted_papers.html
0
13
8
Quick post to close out the week - in our newest preprint,
@julian-stamp.bsky.social
scales the marginal epistasis test to work on biobanks! The key is that using trait-specific information to induce sparsity in the modeled gene-interactions greatly improves both runtime and power
add a skeleton here at some point
about 1 year ago
0
4
1
reposted by
Lorin Crawford
AI x Bio Discovery
about 1 year ago
Evaluating the role of pre-training dataset size and diversity on single-cell foundation model performance
[new]
0
2
1
reposted by
Lorin Crawford
bioRxiv Bioinfo
about 1 year ago
Rethinking cancer drug synergy prediction: a call for standardization in machine learning applications
https://www.biorxiv.org/content/10.1101/2024.12.24.630216v1
0
3
2
In our newest preprint, we show that simply increasing the size of pre-training datasets doesn't necessarily improve the performance of single-cell foundation models on downstream tasks. Really proud of
@alandenadel.bsky.social
for leading this effort! See his thread below for more details👇
add a skeleton here at some point
about 1 year ago
1
5
2
reposted by
Lorin Crawford
Flora Rutaganira
over 1 year ago
This starter pack is in honor of Assistant Professor Antentor Hinton Jr (AJ) at Vanderbilt School of Medicine - Basic Sciences. AJ curated the first 100 Inspiring Black Scientists in America list and in collaboration with the Community of Scholars expanded this list to the 1000.
go.bsky.app/DsJrwR
add a skeleton here at some point
23
481
203
reposted by
Lorin Crawford
Neil Tenenholtz
over 1 year ago
Looking to find your MSR colleagues? I put together a starter pack. Msg me if I missed you (or you just joined).
go.bsky.app/NxtpELZ
add a skeleton here at some point
6
15
7
In the spirit of trying to be more active on here... the deadline for our BioML internship is tomorrow 11/22. If you are currently a PhD student please consider applying --- all we need is a CV and research statement!
jobs.careers.microsoft.com/global/en/jo...
loading . . .
Search Jobs | Microsoft Careers
https://jobs.careers.microsoft.com/global/en/job/1777414/Research-Intern---Machine-Learning-for-Biology-and-Healthcare
over 1 year ago
1
6
4
you reached the end!!
feeds!
log in