Sign in

Paul Thomas

@pgtimmune.bsky.social
3.1K followers 630 following 97 posts

Division of Vaccine and Infectious Disease, Fred Hutchinson Cancer Center TCRs, influenza virus, anti-tumor immunity, books, dogs, Venice

PostsRepliesMedia
Paul Thomas @pgtimmune.bsky.social · 04/09/2026
This work was led by @asya-minervina.bsky.social with @pogorely.bsky.social, Phil Bradley, @sschattgen.bsky.social, Koshlan Mayer-Blackwell, Andrew Fiore-Gartland, Thierry Mora, Aleks Walczak, and Ricky Tirtakasuma. Our first from the @fredhutch.org lab since our move in late 2025!
130
Paul Thomas @pgtimmune.bsky.social · 04/09/2026
There is also a practical opportunity. Starting from a single TCR against a fibrolamellar carcinoma neoantigen, partner chain exchange identified additional neoantigen reactive TCRs with reduced or absent recognition of the corresponding self peptide.
120
Paul Thomas @pgtimmune.bsky.social · 04/09/2026
Structure helped explain part of the story. More promiscuous fixed chains tended to make more predicted contacts with pMHC, leaving greater latitude for their partner chain. But even current structural models could not fully resolve all near identical binder and nonbinder pairs.
100
Paul Thomas @pgtimmune.bsky.social · 04/09/2026
Adding these close negative examples improved TCR specificity classification substantially in our tests. More broadly, the results argue that TCR recognition is a property of the paired receptor, not simply the sum of two independent chain scores.
110
Paul Thomas @pgtimmune.bsky.social · 04/09/2026
Our matched binders and nonbinders create a stringent test set: receptors that differ only in their partner chain & even by one aa, yet have opposite recognition outcomes. Across 10 targets, we recovered more than 70,000 epitope specific TCRs and millions of “close negatives”
110
Paul Thomas @pgtimmune.bsky.social · 04/09/2026
This matters for TCR specificity prediction. A TCR containing a strong antigen motif in one chain can still be nonbinding when paired w/ the wrong partner. Current approaches that lean heavily on single chain similarity can therefore call many close nonbinders positive
110
Paul Thomas @pgtimmune.bsky.social · 04/09/2026
This reveals an asymmetric logic of recognition. For some targets, one chain is relatively promiscuous; it can retain specificity with many partners. For others, the compatible partner is highly constrained. This range spans more than three orders of magnitude.
100
Paul Thomas @pgtimmune.bsky.social · 04/09/2026
The second, and more important, surprise: specificity is hard and wildly variable. Depending on the epitope, 0.06% to 10% of partner chains preserved recognition meaning 90% to 99.9% of candidate partners failed.
100
Paul Thomas @pgtimmune.bsky.social · 04/09/2026
The first surprise was that pairing itself is easy. More than 94% of tested alpha beta combinations formed stable receptors on the cell surface. But surface expression is not the same as antigen recognition.
110
Paul Thomas @pgtimmune.bsky.social · 04/09/2026
We developed a way to hold one TCR chain constant, then let it pair with the enormous natural repertoire of partner chains from healthy donor T cells. In effect, a fixed chain becomes a biological search engine for compatible partners.
110
Paul Thomas @pgtimmune.bsky.social · 04/09/2026
T cells recognize peptide MHC through paired alpha beta T cell receptors. We often find strong antigen associated motifs in one chain, but can one chain alone tell us what a T cell recognizes?
110
Paul Thomas @pgtimmune.bsky.social · 04/09/2026
TCR specificity prediction is advancing rapidly. The central question is not only which TCR sequence recognizes an antigen, but also which TCR pairings make recognition possible & where is the binding boundary? Our new paper tackles these questions. 🧵 www.science.org/doi/10.1126/...
science.org
Uneven TCR chain pairing constraints govern epitope recognition
Combinatorial pairing of independently recombined T cell receptor (TCR) α- and β-chains is central to diversifying the TCR repertoire. Although sequence motifs in one chain correlate with epitope reco...
22210
Reposted by Paul Thomas
Fred Hutch @fredhutch.org · 03/09/2026
“We know both chains are critical for what the TCR sees,” said @pgtimmune.bsky.social, the senior author on new work in the journal @science.org that lays out some of the rules that govern the dance of the TCR chains. “But how much play do we have between the two chains?”
science.org
Uneven TCR chain pairing constraints govern epitope recognition
Combinatorial pairing of independently recombined T cell receptor (TCR) α- and β-chains is central to diversifying the TCR repertoire. Although sequence motifs in one chain correlate with epitope reco...
162
Reposted by Paul Thomas
Joe Sabatino @jsabatino37.bsky.social · 05/02/2026
Overjoyed to share our new work exploring the antigen specificity of CSF-expanded CD8+ T cells in #multiplesclerosis #EBV in @natimmunol.nature.com #immunology 🧪🧵1/ www.nature.com/articles/s41...
nature.com
Antigen specificity of clonally enriched CD8+ T cells in multiple sclerosis - Nature Immunology
Sabatino and colleagues examine expanded CD8+ T cell clonotypes from a small cohort of multiple sclerosis patients. They identified several cognate peptide epitopes that derive from Epstein–Barr virus...
85922
Paul Thomas @pgtimmune.bsky.social · 10/03/2026
It's this Thursday, March 12 at 2 pm ET/11 am PT! Mark your calendars! Hope you versi-can make it!
010
Paul Thomas @pgtimmune.bsky.social · 10/03/2026
This is an ATS @atscommunity.bsky.social RCMB journal club discussion of our paper in the red journal @ajrcmb.bsky.social by Tim Flerlage and David Boyd: Integrated Longitudinal Transcriptomic & Proteomic Analysis of the Murine Lung Response to Influenza A Virus academic.oup.com/ajrcmb/advan...
academic.oup.com
Integrated Longitudinal Transcriptomic and Proteomic Analysis of the Murine Lung Response to Influenza A Virus
Abstract. Lung injury caused by influenza is a leading cause of respiratory infection-related morbidity and mortality worldwide. In its severe form, influe
130
Paul Thomas @pgtimmune.bsky.social · 10/03/2026
Are you a fan of #fibroblasts? Do you think about the lung ECM in your quiet moments? Then join David Boyd & a panel discussion as he presents “Integrated Longitudinal Transcriptomic & Proteomic Analysis of the Murine Lung Response to Influenza A Virus" t.co/2IdbWJVNDQ
t.co
https://thoracic.zoom.us/webinar/register/WN_KiZv45vNRe6bpnnXXJG9nQ
152
Reposted by Paul Thomas
Harry Wallop @harrywallop.co.uk · 02/12/2025
Great culture can save lives. Literally. Amazing letter in today’s @thetimes.com about Tom Stoppard
144117814045
Reposted by Paul Thomas
Mikhail Pogorelyy @pogorely.bsky.social · 25/11/2025
Excited to see our TIRTL-seq method published in Nature Methods www.nature.com/articles/s41...! Some updates after the revision: (1/4)
nature.com
TIRTL-seq: deep, quantitative and affordable paired TCR repertoire sequencing - Nature Methods
TIRTL-seq is a high-throughput method for paired T cell receptor sequencing at the cohort scale.
1197
Paul Thomas @pgtimmune.bsky.social · 25/11/2025
Thanks Sander!
000
Reposted by Paul Thomas
Waggoner Lab @labwaggoner.bsky.social · 25/11/2025
A therapeutic peptide vaccine for fibrolamellar hepatocellular carcinoma @naturemedicine.bsky.social @markyarchoan.bsky.social @hopkinspress.bsky.social @pgtimmune.bsky.social @fredhutch.org @stjude.bsky.social www.nature.com/articles/s41...
051
Reposted by Paul Thomas
Waggoner Lab @labwaggoner.bsky.social · 25/11/2025
@natmethods.nature.com TIRTL-seq: deep, quantitative and affordable paired TCR repertoire sequencing www.nature.com/articles/s41... @pogorely.bsky.social @fredhutch.org @stjude.bsky.social @pgtimmune.bsky.social
0102
Reposted by Paul Thomas
Sylvain Simon @sylvainsimon.bsky.social · 25/11/2025
We are hiring! We have a postdoc position opened @fredhutch.org to join our team to design next gen cellular therapies for cancer. Heavy focus on T cell signaling and understanding how synthetic receptors instruct T cells to fight cancer.
1119
Paul Thomas @pgtimmune.bsky.social · 24/11/2025
FLC work led by @markyarchoan.bsky.social and TIRTL by @pogorely.bsky.social w/ Marina Baretti Allison Kirk, Jeremy Crawford, Samir Adhikari, @nick-clark-bioinfo.bsky.social, Anastasia Minervina, David Brice, Stefan Schattgen, Zeal Kamdar, & Balaji Sundararaman, Support Fibrolamellar Cancer Found.
131
Paul Thomas @pgtimmune.bsky.social · 24/11/2025
If you'd like to try TIRTL, we've got two detailed protocols online, one for the 384-well format (which requires some small volume liquid handlers) and one for a 96-well format which can be done in any lab. protocols.io/view/tirtl-s... and www.protocols.io/view/tirtl-s...
protocols.io
150
Paul Thomas @pgtimmune.bsky.social · 24/11/2025
Together, these studies lay groundwork for rational design of epitope-targeted immunotherapies and demonstrate how high-resolution immunomonitoring can guide next-generation T cell therapies, including public motif-driven adoptive cell transfer.
110
Paul Thomas @pgtimmune.bsky.social · 24/11/2025
Combining vaccine-induced epitope-specific functional data with TIRTL-Seq repertoire analytics produced a powerful stratification model for clinical response, surpassing standard immune monitoring.
120
Paul Thomas @pgtimmune.bsky.social · 24/11/2025
Applying TIRTL-Seq in our FLC vaccine cohort revealed expansive, polyclonal tumor-reactive T cell responses, many shared across patients with common HLA types. This approach exposed public TCR motifs rarely seen in tumor immunity, and mapped the functional landscape of response.
120
Paul Thomas @pgtimmune.bsky.social · 24/11/2025
A major challenge is that tumor- and pathogen-specific T cells circulate at extremely low frequencies, often below 1 in 100,000 or even 1 in a million, making them essentially invisible to standard single-cell approaches.
120
Paul Thomas @pgtimmune.bsky.social · 24/11/2025
To truly interrogate immune responses at the clonal level, we needed unprecedented resolution. Enter TIRTL-Seq: an affordable, scalable platform generating millions of paired TCRαβ clones per experiment, enabling thorough quantification and tracking of T cell dynamics.
120
Paul Thomas @pgtimmune.bsky.social · 24/11/2025
In a first-of-its-kind trial, we combined the FLC vaccine with immune checkpoint blockade. Among patients with advanced disease, 75% mounted strong fusion-specific T cell responses, with 25% experiencing partial responses and durable progression-free survival.
120
Paul Thomas @pgtimmune.bsky.social · 24/11/2025
FLC is a rare, aggressive liver cancer lacking effective systemic therapies. Our vaccine targets the recurrent DNAJB1::PRKACA fusion, offering a universal neoantigen to drive anti-tumor immunity. www.nature.com/articles/s41...
nature.com
A therapeutic peptide vaccine for fibrolamellar hepatocellular carcinoma: a phase 1 trial - Nature Medicine
In this phase 1 trial, treatment of patients with fibrolamellar hepatocellular carcinoma with a therapeutic peptide vaccine targeting the fusion kinase DNAJB1–PRKACA, which is the driver of the diseas...
151
Paul Thomas @pgtimmune.bsky.social · 24/11/2025
Two complementary studies from our group out today: a therapeutic peptide vaccine trial for fibrolamellar carcinoma (FLC) & TIRTL-Seq, our robust method for deep quantitative paired TCR sequencing a🧵w/ @markyarchoan.bsky.social & @pogorely.bsky.social www.nature.com/articles/s41...
nature.com
TIRTL-seq: deep, quantitative and affordable paired TCR repertoire sequencing - Nature Methods
TIRTL-seq is a high-throughput method for paired T cell receptor sequencing at the cohort scale.
43413
Paul Thomas @pgtimmune.bsky.social · 07/11/2025
Thank you!
010
Paul Thomas @pgtimmune.bsky.social · 07/11/2025
Thanks!
000
Paul Thomas @pgtimmune.bsky.social · 07/11/2025
Thank you!
000
Paul Thomas @pgtimmune.bsky.social · 07/11/2025
Haha yes, thanks!
110
Paul Thomas @pgtimmune.bsky.social · 07/11/2025
Thanks Kilian!
010
Reposted by Paul Thomas
Armita Nourmohammad @thearmita.bsky.social · 16/10/2025
excited that this paper is finally out in @pnas.org : www.pnas.org/doi/10.1073/... Led by Gian Marco Visani (effort initiated by Michael Pun), fantastic collaboration with @pgtimmune.bsky.social @asya-minervina.bsky.social and Phil Bradley.
pnas.org
T cell receptor specificity landscape revealed through de novo peptide design | PNAS
T cells play a key role in adaptive immunity by mounting specific responses against diverse pathogens. Effective bindings between T cell receptors ...
0169
Paul Thomas @pgtimmune.bsky.social · 07/11/2025
After many rewarding years at @StJudeResearch surrounded by exceptional colleagues & groundbreaking science, I’m delighted to join @fredhutch to push boundaries in viral & cancer immunology. Grateful to St. Jude and looking forward to what’s next at FH! More here: www.fredhutch.org/en/news/cent...
fredhutch.org
Viral immunologist taps ‘incredible potential’ of immune system
New Fred Hutch Cancer Center viral immunologist Dr. Paul Thomas is working to turn the "incredible potential" of the immune system into real-life diagnostic and therapeutic applications. He studies ho...
710520
Reposted by Paul Thomas
Fred Hutch @fredhutch.org · 07/11/2025
Viral immunologist @pgtimmune.bsky.social, who recently joined Fred Hutch's Vaccine and Infectious Diseases Division, aims to harness the "incredible potential" of the immune system to advance diagnostics, vaccines and cancer therapies. bit.ly/3WJVpos
bit.ly
Viral immunologist taps ‘incredible potential’ of immune system
New Fred Hutch Cancer Center viral immunologist Dr. Paul Thomas is working to turn the "incredible potential" of the immune system into real-life diagnostic and therapeutic applications. He studies ho...
0238
Paul Thomas @pgtimmune.bsky.social · 07/11/2025
Thank you Stacey!! They have all kinds of different birds out here, you need to visit.
010
Paul Thomas @pgtimmune.bsky.social · 15/07/2025
This was a collaborative effort between led by Phil Bradley (not on the socials) and @sschattgen, with Kasi Vegesana, @Asya_Minervina, @villanilab, the MGH COVID-19 team, @s_valkiers and many others!
020
Paul Thomas @pgtimmune.bsky.social · 15/07/2025
In short, you can think of MetaCoNGA as our first draft of the human TCR repertoire. Beta code for matching your T cell populations to those from metaCoNGA is available on Github (github.com/phbradley/me...).
github.com
GitHub - phbradley/metaconga: Scripts and files for meta-analysis with conga
Scripts and files for meta-analysis with conga. Contribute to phbradley/metaconga development by creating an account on GitHub.
120
Paul Thomas @pgtimmune.bsky.social · 15/07/2025
Alternatively, maybe you’ve matched TCRs and GEX to a newly curated regulatory unconventional population (previously difficult to match w/TCR sequence). These vary substantially across donors & conditions & may be highly predictive of immune states relevant to health and disease.
110
Paul Thomas @pgtimmune.bsky.social · 15/07/2025
Maybe you’ve identified a novel condition-associated population and you want to see where it falls–is it a conventional epitope specific response? If so, we might be able to tell you the pathogen, the epitope, or the HLA-restriction (or all 3)?
111
Paul Thomas @pgtimmune.bsky.social · 15/07/2025
Finally, can we put these two analyses together to make a useful tool for the field? We introduce a mapping tool that allows you to take a new data set and match it to the classifications we’ve defined in MetaCoNGA. This has many uses-
110
Paul Thomas @pgtimmune.bsky.social · 15/07/2025
For each population we define a TCR motif and GEX profile, curating known natural Treg, NKT, ILTCK and similar populations, and several novel populations that we can isolate with similar resolution. The result displays the breadth of the unconventional T cell kingdom.
111
Paul Thomas @pgtimmune.bsky.social · 15/07/2025
What do these represent? Lots of known unconventional T cell subsets (MAITs, NKTs, various thymic developmental subsets, KIR+ CD8s, and Tregs) and lots of unknown discrete unconventional populations.
111
Paul Thomas @pgtimmune.bsky.social · 15/07/2025
The idea here is that within GEX space, we look for regions where a subset of neighbors have strongly statistically biased usage of specific TCR amino acids (in no particular order). We find a lot of these neighborhoods! (Over 70K for CD8 & 50K for CD4).
111