Odysseas Vavourakis
@odyv.bsky.social
Generative Antibody Design at Oxford | ovavourakis.github.io | 🇬🇧🇩🇪🇬🇷(🇪🇸) he/him
Thank you for pointing this out. This was due to hitch in our update pipeline; ANARCI seems to number the sequence fine. This entry has now been corrected.
October 13, 2025 at 4:08 PM
Thank you for pointing this out. This was due to hitch in our update pipeline; ANARCI seems to number the sequence fine. This entry has now been corrected.
Huge thanks 🙌 to my fellow members of @opig.stats.ox.ac.uk:
- our lead author Alex Greenshields-Watson
- my co-authors Fabian Spoendlin and @mcagiada.bsky.social
- and our extraordinary P.I. Charlotte Deane!
Have questions or thoughts? Let’s discuss! 🧬
- our lead author Alex Greenshields-Watson
- my co-authors Fabian Spoendlin and @mcagiada.bsky.social
- and our extraordinary P.I. Charlotte Deane!
Have questions or thoughts? Let’s discuss! 🧬
January 27, 2025 at 1:29 AM
Huge thanks 🙌 to my fellow members of @opig.stats.ox.ac.uk:
- our lead author Alex Greenshields-Watson
- my co-authors Fabian Spoendlin and @mcagiada.bsky.social
- and our extraordinary P.I. Charlotte Deane!
Have questions or thoughts? Let’s discuss! 🧬
- our lead author Alex Greenshields-Watson
- my co-authors Fabian Spoendlin and @mcagiada.bsky.social
- and our extraordinary P.I. Charlotte Deane!
Have questions or thoughts? Let’s discuss! 🧬
The future of antibody design is bright, and we’re excited to contribute to it! 🌟
Check out the paper for full details!
Check out the paper for full details!
Challenges and compromises: Predicting unbound antibody structures with deep learning
Therapeutic antibodies are manufactured, stored and administered in the free state; this makes understanding the unbound form key to designing and imp…
www.sciencedirect.com
January 27, 2025 at 1:29 AM
The future of antibody design is bright, and we’re excited to contribute to it! 🌟
Check out the paper for full details!
Check out the paper for full details!
They also give rise to probabilistic metrics (e.g. conformational likelihoods) that could better reflect state occupancies and outperform current metrics as ranking and filtering criteria.
Plus, generative models open the door to robust, antigen-conditional de novo design. 🚀
7/
Plus, generative models open the door to robust, antigen-conditional de novo design. 🚀
7/
January 27, 2025 at 1:29 AM
They also give rise to probabilistic metrics (e.g. conformational likelihoods) that could better reflect state occupancies and outperform current metrics as ranking and filtering criteria.
Plus, generative models open the door to robust, antigen-conditional de novo design. 🚀
7/
Plus, generative models open the door to robust, antigen-conditional de novo design. 🚀
7/
We also suggest generative approaches (like diffusion or flow matching) can help!
Here’s why:
• They target conformational distributions directly as the learning objective.
• They sample these distributions efficiently.
6/
Here’s why:
• They target conformational distributions directly as the learning objective.
• They sample these distributions efficiently.
6/
January 27, 2025 at 1:29 AM
We also suggest generative approaches (like diffusion or flow matching) can help!
Here’s why:
• They target conformational distributions directly as the learning objective.
• They sample these distributions efficiently.
6/
Here’s why:
• They target conformational distributions directly as the learning objective.
• They sample these distributions efficiently.
6/
We call for:
🧠 More ML-grade unbound data for training predictors,
✅ Better methods to rank/QC structure predictions + estimate uncertainty,
🔄 Improved flexibility/ensemble predictions,
🔬 Carrying multiple conformations into downstream analyses.
5/
🧠 More ML-grade unbound data for training predictors,
✅ Better methods to rank/QC structure predictions + estimate uncertainty,
🔄 Improved flexibility/ensemble predictions,
🔬 Carrying multiple conformations into downstream analyses.
5/
January 27, 2025 at 1:29 AM
We call for:
🧠 More ML-grade unbound data for training predictors,
✅ Better methods to rank/QC structure predictions + estimate uncertainty,
🔄 Improved flexibility/ensemble predictions,
🔬 Carrying multiple conformations into downstream analyses.
5/
🧠 More ML-grade unbound data for training predictors,
✅ Better methods to rank/QC structure predictions + estimate uncertainty,
🔄 Improved flexibility/ensemble predictions,
🔬 Carrying multiple conformations into downstream analyses.
5/
In other words, designing better-targeted, more reliable antibodies demands better handling of multiple conformations!
Our paper highlights these challenges, reviews current antibody structure predictors (e.g. AF3, ESM3, ABodyBuilder3), and proposes key directions for progress.
4/
Our paper highlights these challenges, reviews current antibody structure predictors (e.g. AF3, ESM3, ABodyBuilder3), and proposes key directions for progress.
4/
January 27, 2025 at 1:29 AM
In other words, designing better-targeted, more reliable antibodies demands better handling of multiple conformations!
Our paper highlights these challenges, reviews current antibody structure predictors (e.g. AF3, ESM3, ABodyBuilder3), and proposes key directions for progress.
4/
Our paper highlights these challenges, reviews current antibody structure predictors (e.g. AF3, ESM3, ABodyBuilder3), and proposes key directions for progress.
4/
Worse, this conformational heterogeneity directly affects antibody function!
• Entropic contributions influence binding and affinity (ΔG=ΔH–TΔS).
• Flexibility impacts many therapeutic traits.
• Flexibility could even be exploited—e.g., pH-sensitive antibodies that “switch on” inside tumours! 🧪
3/
• Entropic contributions influence binding and affinity (ΔG=ΔH–TΔS).
• Flexibility impacts many therapeutic traits.
• Flexibility could even be exploited—e.g., pH-sensitive antibodies that “switch on” inside tumours! 🧪
3/
January 27, 2025 at 1:29 AM
Worse, this conformational heterogeneity directly affects antibody function!
• Entropic contributions influence binding and affinity (ΔG=ΔH–TΔS).
• Flexibility impacts many therapeutic traits.
• Flexibility could even be exploited—e.g., pH-sensitive antibodies that “switch on” inside tumours! 🧪
3/
• Entropic contributions influence binding and affinity (ΔG=ΔH–TΔS).
• Flexibility impacts many therapeutic traits.
• Flexibility could even be exploited—e.g., pH-sensitive antibodies that “switch on” inside tumours! 🧪
3/
Therapeutic antibodies are manufactured, stored, and administered in their free (unbound) state.
So predicting that conformation is crucial! It’s also hard:
1️⃣ Most antibody structures in the PDB are bound forms, leaving little unbound data.
2️⃣ CDR loops are flexible—literal moving targets!
2/
So predicting that conformation is crucial! It’s also hard:
1️⃣ Most antibody structures in the PDB are bound forms, leaving little unbound data.
2️⃣ CDR loops are flexible—literal moving targets!
2/
January 27, 2025 at 1:29 AM
Therapeutic antibodies are manufactured, stored, and administered in their free (unbound) state.
So predicting that conformation is crucial! It’s also hard:
1️⃣ Most antibody structures in the PDB are bound forms, leaving little unbound data.
2️⃣ CDR loops are flexible—literal moving targets!
2/
So predicting that conformation is crucial! It’s also hard:
1️⃣ Most antibody structures in the PDB are bound forms, leaving little unbound data.
2️⃣ CDR loops are flexible—literal moving targets!
2/