arXiv subjects

Johan Malmström

Publications and source records attributed to Johan Malmström.

3 recordsLinked to original sources

arXiv2026

Controllable protein design with particle-based Feynman-Kac steering

Proteins underpin most biological function, and the ability to design them with tailored structures and properties is central to advances in biotechnology. Diffusion-based generative models have emerged as powerful tools for protein design, but steering them toward proteins with specified properties remains challenging. The Feynman-Kac (FK) framework provides a principled way to guide diffusion models using user-defined rewards. In this paper, we enable FK-based steering of RFdiffusion through the development of guiding potentials that leverage ProteinMPNN and structural relaxation to guide the diffusion process towards desired properties. We show that steering can be used to consistently improve predicted interface energetics and increase binder designability by $89.5\%$. Together, these results establish that diffusion-based protein design can be effectively steered toward arbitrary, non-differentiable objectives, providing a model-independent framework for controllable protein generation.

cs.LG↗

arXiv2026

Deep learning-guided evolutionary optimization for protein design

Designing novel proteins with desired characteristics remains a significant challenge due to the large sequence space and the complexity of sequence-function relationships. Efficient exploration of this space to identify sequences that meet specific design criteria is crucial for advancing therapeutics and biotechnology. Here, we present BoGA (Bayesian Optimization Genetic Algorithm), a framework that combines evolutionary search with Bayesian optimization to efficiently navigate the sequence space. By integrating a genetic algorithm as a stochastic proposal generator within a surrogate modeling loop, BoGA prioritizes candidates based on prior evaluations and surrogate model predictions, enabling data-efficient optimization. We demonstrate the utility of BoGA through benchmarking on sequence and structure design tasks, followed by its application in designing peptide binders against pneumolysin, a key virulence factor of \textit{Streptococcus pneumoniae}. BoGA accelerates the discovery of high-confidence binders, demonstrating the potential for efficient protein design across diverse objectives. The algorithm is implemented within the BoPep suite and is available under an MIT license at \href{https://github.com/ErikHartman/bopep}{GitHub}.

cs.LG↗

arXiv2017

On the Accuracy of Equivalent Antenna Representations

The accuracy of two equivalent antenna representations, near-field sources and far-field sources, are evaluated for an antenna installed on a simplified platform in a series of case studies using different configurations of equivalent antenna representations. The accuracy is evaluated in terms of installed far-fields and surface currents on the platform. The results show large variations between configurations. The root-mean-square installed far-field error is 4.4% for the most accurate equivalent representation. When using far-field sources, the design parameters have a large influence of the achieved accuracy. There is also a varying accuracy depending on the type of numerical method used. Based on the results, some recommendations on the choice of sub-domain for the equivalent antenna representation are given. In industrial antenna applications, the accuracy in determining e.g. installed far-fields and antenna isolation on large platforms are critical. Equivalent representations can reduce the fine-detail complexity of antennas and thus give an efficient numerical descriptions to be used in large-scale simulations. The results in this paper can be used as a guideline by antenna designers or system engineers when using equivalent sources.

physics.class-ph↗