Alexander Ille

Excited to share our work on modeling the human protein interactome in 3D. This currently includes 1,394 interaction structures predicted with Boltz-2. Grateful for the open-access resources making this possible!

Preprint: doi.org/10.1101/2025...

The prediction of protein conformational dynamics from sequence is expected to have broad utility across biological and medical research. Huge thank you to my co-authors for contributing to this piece!

Importantly, there is an abundance of sequence data and increasing availability of conformationally-sensitive structure data which may be leveraged for AI/ML model training.

Based on the above conjecture, AI/ML models may be developed to predict a protein’s conformational dynamics from its sequence. We provide a conceptual architecture for such a model, drawing from previous work which provides encouraging precedent.

In the current piece, we consider that biological sequence encodes not only static 3D structure, but also conformational dynamics. This is important because inside of a cell, things aren’t static and motionless, they are constantly moving around.

Later on, Christian Anfinsen proposed that “the native conformation is determined by the totality of interatomic interactions and hence by the amino acid sequence” (Anfinsen, 1973).

On the relationship between sequence and structure, Francis Crick speculated that “folding is simply a function of the order of the amino acids” (Crick, 1958).

Excited to share a new article I co-authored on the relationship between biological sequence and structure! We look back to historic hypotheses, and look forward to how AI/ML models may be developed to predict protein conformational dynamics. Paper: doi.org/10.1063/4.00...

Hemoglobin, the biomolecule that breathes, modeled with Boltz-2.

With the recent release of #Boltz2, I gave it a try for modeling hemoglobin. Hemoglobin structures were likely included for training, and this was just a fun test, but it’s awesome to see this modeling in action!

Truly remarkable work. Looking forward to using it!

Dynamic insight into DNA → RNA → protein. Great work!

PDB101: Paper Published: Visualizing and analyzing 3D biomolecular structures using Mol* PDB-101: Training, Outreach, and Education portal of RCSB PDB

New RCSB PDB publication--
Read how Mol* can be used to visualize and analyze the influenza A H5N1 virus proteome
pdb101.rcsb.org/news...

Computational capacity of life in relation to the universe The discovery of life processing with UV-excited qubits supports a conjecture relative to the computing capacity of the universe.

Computational capacity of life in relation to the universe | Science Advances www.science.org/doi/10.1126/...

AlphaFold can be used to predict the oligomeric states of proteins Homooligomerisation is a prevalent and important process that many proteins undergo to form the quaternary structure required to carry out their biological functions. However, determining the oligomer...

Creative use of AlphaFold for predicting oligomeric states of proteins, nice work.
www.biorxiv.org/content/10.1...

Predictomes, a classifier-curated database of AlphaFold-modeled protein-protein interactions Schmid and Walter train a classifier that discerns functionally relevant structure predictions in proteome-wide protein-protein interaction (PPI) screens using AlphaFold-Multimer, and they use this co...

Predictomes (Schmid & Walter, 2025). This looks very useful!
www.cell.com/molecular-ce...

Never gets old. #Discovery