Frederick "Erick" Matsen
@matsen.bsky.social
I ♥ evolution, immunology, math, & computers. Professor at Fred Hutch & Investigator at HHMI. http://matsen.fredhutch.org/
The Mahan postdoctoral fellowship offers 21 months of support to develop your own research with Fred Hutch computational biology faculty-- lots of excellent labs to choose from!
Apply: apply.interfolio.com/172697
Faculty: www.fredhutch.org/en/research...
Apply: apply.interfolio.com/172697
Faculty: www.fredhutch.org/en/research...
Herbold Computational Biology Faculty & Labs
www.fredhutch.org
October 19, 2025 at 5:40 PM
The Mahan postdoctoral fellowship offers 21 months of support to develop your own research with Fred Hutch computational biology faculty-- lots of excellent labs to choose from!
Apply: apply.interfolio.com/172697
Faculty: www.fredhutch.org/en/research...
Apply: apply.interfolio.com/172697
Faculty: www.fredhutch.org/en/research...
The final version of our transformer-based model of natural selection has come out in MBE. I hope some molecular evolution researchers find this interesting & useful as a way to express richer models of natural selection. doi.org/10.1093/mol... (short 🧵)
September 24, 2025 at 10:24 PM
The final version of our transformer-based model of natural selection has come out in MBE. I hope some molecular evolution researchers find this interesting & useful as a way to express richer models of natural selection. doi.org/10.1093/mol... (short 🧵)
Reposted by Frederick "Erick" Matsen
We are looking for an #AIEngineer to help build protein language models that capture evolutionary constraints with @matsen.bsky.social and @jbloomlab.bsky.social at #AI@HHMI @hhmijanelia.bsky.social
hhmi.wd1.myworkdayjobs.com/en-US/Extern...
hhmi.wd1.myworkdayjobs.com/en-US/Extern...
AI Engineer - Evolutionary Protein Language Models
Primary Work Address: 19700 Helix Drive, Ashburn, VA, 20147 Current HHMI Employees, click here to apply via your Workday account. Intro: AI@HHMI: HHMI is investing $500 million over the next 10 years ...
hhmi.wd1.myworkdayjobs.com
September 18, 2025 at 4:50 PM
We are looking for an #AIEngineer to help build protein language models that capture evolutionary constraints with @matsen.bsky.social and @jbloomlab.bsky.social at #AI@HHMI @hhmijanelia.bsky.social
hhmi.wd1.myworkdayjobs.com/en-US/Extern...
hhmi.wd1.myworkdayjobs.com/en-US/Extern...
The final version of our "Thrifty" paper is up now: elifesciences.org/articles/10... .
We were motivated to fit wide-context mutation models based on previous analyses showing "mesoscale" effects and a position-specific effect. But, how to avoid exploding the number of parameters? 🧵
We were motivated to fit wide-context mutation models based on previous analyses showing "mesoscale" effects and a position-specific effect. But, how to avoid exploding the number of parameters? 🧵
Thrifty wide-context models of B cell receptor somatic hypermutation
Convolutional embedding models efficiently capture wide sequence context in antibody somatic hypermutation, avoiding exponential k-mer parameter scaling and eliminating the need for per-site modeling.
elifesciences.org
September 18, 2025 at 10:46 PM
The final version of our "Thrifty" paper is up now: elifesciences.org/articles/10... .
We were motivated to fit wide-context mutation models based on previous analyses showing "mesoscale" effects and a position-specific effect. But, how to avoid exploding the number of parameters? 🧵
We were motivated to fit wide-context mutation models based on previous analyses showing "mesoscale" effects and a position-specific effect. But, how to avoid exploding the number of parameters? 🧵
Is an idea likely to advance the field?
Our "preflight check" exercise provides a structured approach for thinking through computational biology research projects.
github.com/matsengrp/pr...
Thanks to @sdwfrost.bsky.social for the core idea!
Our "preflight check" exercise provides a structured approach for thinking through computational biology research projects.
github.com/matsengrp/pr...
Thanks to @sdwfrost.bsky.social for the core idea!
GitHub - matsengrp/preflight
Contribute to matsengrp/preflight development by creating an account on GitHub.
github.com
September 16, 2025 at 2:06 PM
Is an idea likely to advance the field?
Our "preflight check" exercise provides a structured approach for thinking through computational biology research projects.
github.com/matsengrp/pr...
Thanks to @sdwfrost.bsky.social for the core idea!
Our "preflight check" exercise provides a structured approach for thinking through computational biology research projects.
github.com/matsengrp/pr...
Thanks to @sdwfrost.bsky.social for the core idea!
Interested in doing a PhD or postdoc in our group? Here is a letter to you: matsen.group/general/202...
We are ready to recruit a trainee who can help develop the next generation of our transformer-based models of natural selection. See the "joining" tab of our website for details.
We are ready to recruit a trainee who can help develop the next generation of our transformer-based models of natural selection. See the "joining" tab of our website for details.
Dear future trainee:
Let's have fun, work hard, and feel lucky that our job is to expand the boundary of knowledge.
matsen.group
August 21, 2025 at 9:47 AM
Interested in doing a PhD or postdoc in our group? Here is a letter to you: matsen.group/general/202...
We are ready to recruit a trainee who can help develop the next generation of our transformer-based models of natural selection. See the "joining" tab of our website for details.
We are ready to recruit a trainee who can help develop the next generation of our transformer-based models of natural selection. See the "joining" tab of our website for details.
Open bioinformatics position on next-generation protein evolution models! Join HHMI's AI initiative at Janelia Farm, Virginia, (an amazing place) and work closely with our team. Help us build the future! 🧬 + 🤖 = ❤️
hhmi.wd1.myworkdayjobs.com/en-US/Exter...
hhmi.wd1.myworkdayjobs.com/en-US/Exter...
August 20, 2025 at 11:17 AM
Open bioinformatics position on next-generation protein evolution models! Join HHMI's AI initiative at Janelia Farm, Virginia, (an amazing place) and work closely with our team. Help us build the future! 🧬 + 🤖 = ❤️
hhmi.wd1.myworkdayjobs.com/en-US/Exter...
hhmi.wd1.myworkdayjobs.com/en-US/Exter...
Why does selection feel so weak relative to mutation in affinity maturation? A new blog post giving three perspectives, including our new transformer-based model of natural selection on antibodies: matsen.group/general/202...
The term 'affinity maturation' understates the influence of somatic hypermutation
Three recent papers quantify how nucleotide-level mutation processes drive antibody evolution.
matsen.group
August 19, 2025 at 5:17 PM
Why does selection feel so weak relative to mutation in affinity maturation? A new blog post giving three perspectives, including our new transformer-based model of natural selection on antibodies: matsen.group/general/202...
Reposted by Frederick "Erick" Matsen
In a new preprint we use deep learning on lineage trees to infer the functional form of the relationship between affinity and fitness that controls antibody evolution in germinal centers: arxiv.org/abs/2508.09871 🧵
Inference of germinal center evolutionary dynamics via simulation-based deep learning
B cells and the antibodies they produce are vital to health and survival, motivating research on the details of the mutational and evolutionary processes in the germinal centers (GC) from which mature...
arxiv.org
August 16, 2025 at 10:56 PM
In a new preprint we use deep learning on lineage trees to infer the functional form of the relationship between affinity and fitness that controls antibody evolution in germinal centers: arxiv.org/abs/2508.09871 🧵
Here are some useful subagents we've developed for Claude Code.
github.com/matsengrp/c...
(description in README if you don't know what I'm talking about)
Example uses:
github.com/matsengrp/c...
(description in README if you don't know what I'm talking about)
Example uses:
August 14, 2025 at 1:40 PM
Here are some useful subagents we've developed for Claude Code.
github.com/matsengrp/c...
(description in README if you don't know what I'm talking about)
Example uses:
github.com/matsengrp/c...
(description in README if you don't know what I'm talking about)
Example uses:
Motivated by wanting Claude Code to read papers, and something to fill PDF forms for kid summer camps, I vibe-coded github.com/matsengrp/p...
Perhaps you will find it useful!
Perhaps you will find it useful!
GitHub - matsengrp/pdf-navigator-mcp: Comprehensive MCP server for PDF reading, navigation, and text search
Comprehensive MCP server for PDF reading, navigation, and text search - matsengrp/pdf-navigator-mcp
github.com
August 3, 2025 at 2:17 PM
Motivated by wanting Claude Code to read papers, and something to fill PDF forms for kid summer camps, I vibe-coded github.com/matsengrp/p...
Perhaps you will find it useful!
Perhaps you will find it useful!
Reposted by Frederick "Erick" Matsen
Go Maggie! @magdalenarussell.bsky.social gets UW's Distinguished Dissertation Award for her PhD work with @matsen.bsky.social "Inferring mechanisms of V(D)J recombination using statistical inference on high-throughput immune repertoire data". 🏆
Congratulations to Maggie Russell on receiving the UW Graduate School's Distinguished Dissertation Award!
July 3, 2025 at 10:19 PM
Go Maggie! @magdalenarussell.bsky.social gets UW's Distinguished Dissertation Award for her PhD work with @matsen.bsky.social "Inferring mechanisms of V(D)J recombination using statistical inference on high-throughput immune repertoire data". 🏆
Reposted by Frederick "Erick" Matsen
Excited to share my new preprint developed with @matsen.bsky.social, in collaboration with Marius Brusselmans, Luiz Carvalho, @msuchard.bsky.social, and @guybaele.bsky.social, on the biological causes and impacts of tree space ruggedness in phylodynamic inference. 1/
www.biorxiv.org/content/10.1...
www.biorxiv.org/content/10.1...
www.biorxiv.org
June 17, 2025 at 1:25 AM
Excited to share my new preprint developed with @matsen.bsky.social, in collaboration with Marius Brusselmans, Luiz Carvalho, @msuchard.bsky.social, and @guybaele.bsky.social, on the biological causes and impacts of tree space ruggedness in phylodynamic inference. 1/
www.biorxiv.org/content/10.1...
www.biorxiv.org/content/10.1...
Reposted by Frederick "Erick" Matsen
Wanted to highlight our latest preprint--a huge effort by multiple people and labs, but led primarily by @wsdewitt.github.io, Tatsuya Araki, and Ashni Vora, in a very close wet-dry collaboration with @matsen.bsky.social’s lab at the Hutch
www.biorxiv.org/content/10.1...
www.biorxiv.org/content/10.1...
Replaying germinal center evolution on a quantified affinity landscape
Darwinian evolution of immunoglobulin genes within germinal centers (GC) underlies the progressive increase in antibody affinity following antigen exposure. Whereas the mechanics of how competition be...
www.biorxiv.org
June 5, 2025 at 2:28 PM
Wanted to highlight our latest preprint--a huge effort by multiple people and labs, but led primarily by @wsdewitt.github.io, Tatsuya Araki, and Ashni Vora, in a very close wet-dry collaboration with @matsen.bsky.social’s lab at the Hutch
www.biorxiv.org/content/10.1...
www.biorxiv.org/content/10.1...
Working with Claude Code has been transformative. Yes, it does great on new code and refactoring. VSCode integration is 👍.
But the big surprise: it can seamlessly edit notebooks and then view the plots to plan next steps. Time to update my blog post matsen.group/general/202...
But the big surprise: it can seamlessly edit notebooks and then view the plots to plan next steps. Time to update my blog post matsen.group/general/202...
June 4, 2025 at 6:27 PM
Working with Claude Code has been transformative. Yes, it does great on new code and refactoring. VSCode integration is 👍.
But the big surprise: it can seamlessly edit notebooks and then view the plots to plan next steps. Time to update my blog post matsen.group/general/202...
But the big surprise: it can seamlessly edit notebooks and then view the plots to plan next steps. Time to update my blog post matsen.group/general/202...
Many of my favorite collaborators are immigrants who face double threats of funding cuts and the anti-immigrant political climate.
To them and immigrants generally: we need you, and we want you here. I understand if you want to leave but I hope you will stay and reach out.
To them and immigrants generally: we need you, and we want you here. I understand if you want to leave but I hope you will stay and reach out.
May 31, 2025 at 10:12 AM
Many of my favorite collaborators are immigrants who face double threats of funding cuts and the anti-immigrant political climate.
To them and immigrants generally: we need you, and we want you here. I understand if you want to leave but I hope you will stay and reach out.
To them and immigrants generally: we need you, and we want you here. I understand if you want to leave but I hope you will stay and reach out.
I'm excited about AI not as a way to get a big black-box model, but as a way to assist development of probabilistic models that get us to biological mechanism. New blog post: matsen.group/general/2025...
The next five years are going to be amazing for learning biological processes through probabilistic models
Biology is composed of many small interacting processes. We can use AI to develop models accordingly.
matsen.group
May 28, 2025 at 4:58 PM
I'm excited about AI not as a way to get a big black-box model, but as a way to assist development of probabilistic models that get us to biological mechanism. New blog post: matsen.group/general/2025...
New blog post: our recent work to understand the somatic hypermutation process that enables antibodies to incrementally improve.
A story of running into the limitations of deep learning, but still gaining biological insight along the way.
matsen.group/general/2025...
A story of running into the limitations of deep learning, but still gaining biological insight along the way.
matsen.group/general/2025...
Two new approaches to learn about antibody somatic hypermutation
We try to learn about SHM mechanism and do so by directly using mechanistic models, and by exploring deep architectures.
matsen.group
May 20, 2025 at 5:30 PM
New blog post: our recent work to understand the somatic hypermutation process that enables antibodies to incrementally improve.
A story of running into the limitations of deep learning, but still gaining biological insight along the way.
matsen.group/general/2025...
A story of running into the limitations of deep learning, but still gaining biological insight along the way.
matsen.group/general/2025...
I'm excited about this talk in a week. Faruck has been doing coevolution -> structure since before it was cool, e.g. www.pnas.org/doi/abs/10.1...
April 22, 2025 at 4:09 PM
I'm excited about this talk in a week. Faruck has been doing coevolution -> structure since before it was cool, e.g. www.pnas.org/doi/abs/10.1...
Congratulations to Maggie Russell @magdalenarussell.bsky.social for graduating from @mcbseattle.bsky.social with a phenomenal thesis. I will retell the story of her thesis through images from her papers. 🧵
(Here we are with Phil Bradley, who co-advised Maggie.)
(Here we are with Phil Bradley, who co-advised Maggie.)
February 26, 2025 at 12:46 AM
Congratulations to Maggie Russell @magdalenarussell.bsky.social for graduating from @mcbseattle.bsky.social with a phenomenal thesis. I will retell the story of her thesis through images from her papers. 🧵
(Here we are with Phil Bradley, who co-advised Maggie.)
(Here we are with Phil Bradley, who co-advised Maggie.)
Big news from Sebastien Roch and collabs on phylogenetic likelihood function: "... we show that the expected log-likelihood is strongly concave and smooth in a box around the true parameter whose size is independent of both the tree topology and number of leaves."
arxiv.org/abs/2501.17622
arxiv.org/abs/2501.17622
Likelihood landscape of binary latent model on a tree
We study the optimization landscape of maximum likelihood estimation for a binary latent tree model with hidden variables at internal nodes and observed variables at the leaves. This model, known as t...
arxiv.org
February 2, 2025 at 8:57 PM
Big news from Sebastien Roch and collabs on phylogenetic likelihood function: "... we show that the expected log-likelihood is strongly concave and smooth in a box around the true parameter whose size is independent of both the tree topology and number of leaves."
arxiv.org/abs/2501.17622
arxiv.org/abs/2501.17622
Reposted by Frederick "Erick" Matsen
Building on the UShER tree of millions of SARS-CoV-2 genomes maintained by Angie Hinrichs, Hugh Haddox and Georg Angehrn (and others in @matsen.bsky.social lab and @jbloomlab.bsky.social) have looked into how the neutral mutation rate varies along the genome:
[1/N]
www.biorxiv.org/content/10.1...
[1/N]
www.biorxiv.org/content/10.1...
The mutation rate of SARS-CoV-2 is highly variable between sites and is influenced by sequence context, genomic region, and RNA structure
RNA viruses like SARS-CoV-2 have a high mutation rate, which contributes to their rapid evolution. The rate of mutations depends on the mutation type (e.g., A→C, A→G, etc.) and can vary between sites ...
www.biorxiv.org
January 11, 2025 at 5:37 PM
Building on the UShER tree of millions of SARS-CoV-2 genomes maintained by Angie Hinrichs, Hugh Haddox and Georg Angehrn (and others in @matsen.bsky.social lab and @jbloomlab.bsky.social) have looked into how the neutral mutation rate varies along the genome:
[1/N]
www.biorxiv.org/content/10.1...
[1/N]
www.biorxiv.org/content/10.1...
Reposted by Frederick "Erick" Matsen
Large protein language models can learn complex epistatic interactions, but how much does that help with predicting variant effects? In this NeurIPS article, we show that classical independent-sites phylogenetic models can outperform pLMs on this task.
1/7
openreview.net/forum?id=H7m...
1/7
openreview.net/forum?id=H7m...
Ultrafast classical phylogenetic method beats large protein...
Amino acid substitution rate matrices are fundamental to statistical phylogenetics and evolutionary biology. Estimating them typically requires reconstructed trees for massive amounts of aligned...
openreview.net
November 16, 2024 at 8:42 PM
Large protein language models can learn complex epistatic interactions, but how much does that help with predicting variant effects? In this NeurIPS article, we show that classical independent-sites phylogenetic models can outperform pLMs on this task.
1/7
openreview.net/forum?id=H7m...
1/7
openreview.net/forum?id=H7m...
Reposted by Frederick "Erick" Matsen
We are here! All future seminar announcements will be posted here.
November 16, 2024 at 3:50 PM
We are here! All future seminar announcements will be posted here.