arXiv subjects

Mattia G. Gollub

Publications and source records attributed to Mattia G. Gollub.

4 recordsLinked to original sources

arXiv2026

Continuous Variational Synthesis

Biological machine learning was long bottlenecked by the ability to synthesize designed DNA. Variational synthesis models control chemical reactions to physically manufacture quadrillions of designed sequences in DNA. However, training these generative models is challenging: constraints on chemical synthesis can force many parameters into a discrete space, limiting the ability to pre-train and fine-tune. In this article we train ``free'' variational synthesis models using stochastic gradient descent in continuous space, and then discretize with post-training quantization to impose hardware and wetware constraints. This enables variational synthesis models to satisfy stringent reward criteria, while still synthesizing diverse designs, achieving a strictly dominating quality-diversity Pareto frontier. We demonstrate by training variational synthesis models of enzymes, peptides, antibody CDRH3s, and regulatory DNA elements. In silico performance is maintained in vitro.

q-bio.BM↗

arXiv2025

Lifting Biomolecular Data Acquisition

One strategy to scale up ML-driven science is to increase wet lab experiments' information density. We present a method based on a neural extension of compressed sensing to function space. We measure the activity of multiple different molecules simultaneously, rather than individually. Then, we deconvolute the molecule-activity map during model training. Co-design of wet lab experiments and learning algorithms provably leads to orders-of-magnitude gains in information density. We demonstrate on antibodies and cell therapies.

q-bio.BM↗

arXiv2025

Accelerated Learning on Large Scale Screens using Generative Library Models

Biological machine learning is often bottlenecked by a lack of scaled data. One promising route to relieving data bottlenecks is through high throughput screens, which can experimentally test the activity of $10^6-10^{12}$ protein sequences in parallel. In this article, we introduce algorithms to optimize high throughput screens for data creation and model training. We focus on the large scale regime, where dataset sizes are limited by the cost of measurement and sequencing. We show that when active sequences are rare, we maximize information gain if we only collect positive examples of active sequences, i.e. $x$ with $y>0$. We can correct for the missing negative examples using a generative model of the library, producing a consistent and efficient estimate of the true $p(y | x)$. We demonstrate this approach in simulation and on a large scale screen of antibodies. Overall, co-design of experiments and inference lets us accelerate learning dramatically.

stat.ML↗

arXiv2021

eQuilibrator 3.0 -- a platform for the estimation of thermodynamic constants

eQuilibrator (equilibrator.weizmann.ac.il) is a calculator for biochemical equilibrium constants and Gibbs free energies, originally designed as a web-based interface. While the website now counts ${\sim}1000$ distinct monthly users, its design could not accommodate larger compound databases and it lacked an application programming interface (API) for integration in other tools developed by the systems biology community. Here, we report a new python-based package for eQuilibrator, that comes with many new features such as a 50-fold larger compound database, the ability to add novel compound structures, improvements in speed and memory use, and correction for Mg2+ ion concentrations. Moreover, it adds the ability to compute the covariance matrix of the uncertainty between estimates, for which we show the advantages and describe the application in metabolic modeling. We foresee that these improvements will make thermodynamic modeling more accessible and facilitate the integration of eQuilibrator into other software platforms.

q-bio.MN↗