GFlowNets for AI-driven scientific discovery

Moksh, Jain,; Tristan, Deleu,; Hartford, Jason; Liu, Chenghao; Hernández-García, Álex; Bengio, Yoshua

doi:10.1039/d3dd00002h

Cited by 8 publications

(5 citation statements)

References 85 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, Brookes et al ( 2019) built variational autoencoders (VAEs) that generate desirable designs by iteratively improving the VAE and its designs, guided by a design model. Jain et al (2022) used GFlowNets for sequence design, another class of generative models that are trained to output designs that span all modes of some design model's prediction distribution. Linder et al (2020)'s deep exploration networks (DENs) are generative models that are trained to output diverse yet desirable sequences and again use design model predictions to guide the training process.…”

Section: Offline Mbo For Designing Biological Se-mentioning

confidence: 99%

Designing Cell-Type-Specific Promoter Sequences Using Conservative Model-Based Optimization

Reddy,

Geng,

Herschl

et al. 2024

Preprint

View full text Add to dashboard Cite

Gene therapies have the potential to treat disease by delivering therapeutic genetic cargo to disease-associated cells. One limitation to their widespread use is the lack of short regulatory sequences, or promoters, that differentially induce the expression of delivered genetic cargo in target cells, minimizing side effects in other cell types. Such cell-type-specific promoters are difficult to discover using existing methods, requiring either manual curation or access to large datasets of promoter-driven expression from both targeted and untargeted cells. Model-based optimization (MBO) has emerged as an effective method to design biological sequences in an automated manner, and has recently been used in promoter design methods. However, these methods have only been tested using large training datasets that are expensive to collect, and focus on designing promoters for markedly different cell types, overlooking the complexities associated with designing promoters for closely related cell types that share similar regulatory features. Therefore, we introduce a comprehensive framework for utilizing MBO to design promoters in a data-efficient manner, with an emphasis on discovering promoters for similar cell types. We use conservative objective models (COMs) for MBO and highlight practical considerations such as best practices for improving sequence diversity, getting estimates of model uncertainty, and choosing the optimal set of sequences for experimental validation. Using three relatively similar blood cancer cell lines (Jurkat, K562, and THP1), we show that our approach discovers many novel cell-type-specific promoters after experimentally validating the designed sequences. For K562 cells, in particular, we discover a promoter that has 75.85% higher cell-type-specificity than the best promoter from the initial dataset used to train our models.

show abstract

Section: Offline Mbo For Designing Biological Se-mentioning

confidence: 99%

Designing Cell-Type-Specific Promoter Sequences Using Conservative Model-Based Optimization

Reddy,

Geng,

Herschl

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…By relying on the new experimental data, only a limited number of training samples can be expected, confining the choice of ML models to those with a lower number of parameters, such as multilayer perceptrons (MLP) with as few as two layers . A recent advance in the area of active learning is the development of GFlowNets, networks designed to suggest diverse and accurate candidates in a machine–expert loop to accelerate scientific discovery. Studies using such networks have demonstrated their potential for designing small molecules; however, the utility for the design of proteins remains to be reliably demonstrated.…”

Section: Protein Engineering Tasks Solved By Machine Learningmentioning

confidence: 99%

Machine Learning-Guided Protein Engineering

Kouba,

Kohout,

Haddadi

et al. 2023

ACS Catal.

View full text Add to dashboard Cite

Recent progress in engineering highly promising biocatalysts has increasingly involved machine learning methods. These methods leverage existing experimental and simulation data to aid in the discovery and annotation of promising enzymes, as well as in suggesting beneficial mutations for improving known targets. The field of machine learning for protein engineering is gathering steam, driven by recent success stories and notable progress in other areas. It already encompasses ambitious tasks such as understanding and predicting protein structure and function, catalytic efficiency, enantioselectivity, protein dynamics, stability, solubility, aggregation, and more. Nonetheless, the field is still evolving, with many challenges to overcome and questions to address. In this Perspective, we provide an overview of ongoing trends in this domain, highlight recent case studies, and examine the current limitations of machine learning-based methods. We emphasize the crucial importance of thorough experimental validation of emerging models before their use for rational protein design. We present our opinions on the fundamental problems and outline the potential directions for future research.

show abstract

“…The action distribution induced by this EBM in a structured action space is highly multimodal, and sampling from such a high-dimensional distribution is intractable. To this end, we introduce a recently proposed novel and powerful generative model, Generative Flow Networks (GFlowNet) [4,5,33,99], as the efficient diverse policy sampler. GFlowNet can be regarded as amortized Monte-Carlo Markov chains (MCMC), which gradually builds composable environmental actions through the single but trained generative pass of "building blocks (i.e., atomic actions)", so that the final sampled environmental actions obey a given energy-based policy distribution.…”

Section: Introductionmentioning

confidence: 99%

Diverse Policy Optimization for Structured Action Space

Li¹,

Wang²,

Yang³

et al. 2023

Preprint

View full text Add to dashboard Cite

Enhancing the diversity of policies is beneficial for robustness, exploration, and transfer in reinforcement learning (RL). In this paper, we aim to seek diverse policies in an under-explored setting, namely RL tasks with structured action spaces with the two properties of composability and local dependencies. The complex action structure, non-uniform reward landscape, and subtle hyperparameter tuning due to the properties of structured actions prevent existing approaches from scaling well. We propose a simple and effective RL method, Diverse Policy Optimization (DPO), to model the policies in structured action space as the energy-based models (EBM) by following the probabilistic RL framework. A recently proposed novel and powerful generative model, GFlowNet, is introduced as the efficient, diverse EBM-based policy sampler. DPO follows a joint optimization framework: the outer layer uses the diverse policies sampled by the GFlowNet to update the EBM-based policies, which supports the GFlowNet training in the inner layer. Experiments on ATSC and Battle benchmarks demonstrate that DPO can efficiently discover surprisingly diverse policies in challenging scenarios and substantially outperform existing state-of-the-art methods.

show abstract

GFlowNets for AI-driven scientific discovery

Abstract: Tackling the most pressing problems for humanity, such as the climate crisis and the threat of global pandemics, requires accelerating the pace of scientific discovery. While science has traditionally relied...

Cited by 8 publications

References 85 publications

Designing Cell-Type-Specific Promoter Sequences Using Conservative Model-Based Optimization

Designing Cell-Type-Specific Promoter Sequences Using Conservative Model-Based Optimization

Machine Learning-Guided Protein Engineering

Diverse Policy Optimization for Structured Action Space

Contact Info

Product

Resources

About