Biology subjects

Stalmarck, A.

Publications and source records attributed to Stalmarck, A..

3 recordsLinked to original sources

bioRxiv2025

Spontaneous Emergence of Symmetry in a Generative Model of Protein Structure

Generative models are becoming powerful tools for protein design, enabling the creation of novel protein structures and sequences. Recent approaches have shown success using diffusion models and flow matching to sample realistic protein folds. Many methods explicitly incorporate structural biases or constraints--for example, enforcing symmetry during generation--to steer the design process. Here we report the spontaneous emergence of structural symmetry in a transformer-based generative model for proteins, without any symmetry-specific conditioning during training or constraints during generation. In our flow-matching model, a single attention head in an SE(3)-equivariant transformer layer was found to be primarily responsible for the models ability to generate symmetric arrangements of backbone residues across chains, or repeating motifs within a chain. Our results show that protein generative models can learn high-level structural patterns implicitly from training data. This opens new questions about interpretability and control in generative design: understanding how and why a single attention head can govern a complex global property like symmetry may inform future model architectures and help exploit emergent behaviors for better protein engineering.

bioinformatics↗

bioRxiv2025

Detection of PCR chimeras in adaptive immune receptor repertoire sequencing using hidden Markov models

Adaptive Immune Receptor Repertoire sequencing (AIRR-seq) has emerged as a central approach for studying T cell and B cell receptor populations, and is now an important component of studies of autoimmunity, immune responses to pathogens, vaccines, allergens, and cancers, and for antibody discovery. When amplifying the rearranged V(D)J genes encoding antigen receptors, each cycle of the Polymerase Chain Reaction (PCR) can produce spurious "chimeric" hybrids of two or more different template sequences. While the generation of chimeras is well understood in bacterial and viral sequencing, and there are dedicated tools to detect such sequences in bacterial and viral datasets, this is not the case for AIRR-seq. Further, the process that results in immune receptor sequences has domain-specific challenges, such as somatic hypermutation (SHM), and domain-specific opportunities, such as relatively well-known germline gene "reference" sequences. Here we describe CHMMAIRRa, a hidden Markov model for detecting chimeric sequences in AIRR-seq data, that specifically models SHM and incorporates germline reference sequences. We use simulations to characterize the performance of CHMMAIRRa and compare it to existing methods from other domains, we test the effect of PCR conditions on chimerism using IgM libraries generated in this study, and we apply CHMMAIRRa to four published AIRR-seq datasets to show the extent and impact of artifactual chimerism.

bioinformatics↗

bioRxiv2024

The Continuous Language of Protein Structure

Just as language is composed of sublexical tokens that combine to form words, sentences, and paragraphs, protein backbones are composed of sub-structural elements that combine to form helices, sheets, folds, domains, and chains. Autoregressive language models operate on discrete tokens, whereas protein structure is inherently continuous, and generative approaches to protein design have borrowed more from image generation than language modeling. But autoregressive models do not inherently require their inputs and outputs to be discrete. Here we describe a generative autoregressive language model over the continuous space of protein backbones, where the distribution over the placement of each successive amino acid is conditioned on all preceding residues, and can be sampled from one residue after another. We show that this approach can learn to sample diverse and realistic protein chains, opening a new potential avenue for in silico protein design.

bioinformatics↗