FLOP: Tasks for Fitness Landscapes Of Protein wildtypes

Groth, Paul; Richard, Michael; Salomon, Jesper; Tian, Pengfei; Boomsma, Wouter

doi:10.1101/2023.06.21.545880

Cited by 4 publications

(3 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…54,55 Overall, there is also still room to develop better benchmarks for enzyme discovery, to measure the effectiveness of various models and representations. 56 Recently, there has been an explosion in protein structure data from ML-enabled protein structure prediction tools such as AlphaFold2 and others 57−62 and databases of unannotated protein structures. Clustering similar structures is one way to annotate for function.…”

Section: Annotation Of Enzyme Activity Among Known Proteinsmentioning

confidence: 99%

“…EC numbers do not capture a quantitative notion of similarity between reactions, so enzyme activity prediction would benefit from a learned continuous representation of the similarity between activities, where reactions, substrates, and products are numerically encoded. This could resemble current efforts to encode chemical structures and predict the outcomes of reactions in synthetic organic chemistry. − Databases will be useful for the curation and standardization of enzyme reaction data. , Overall, there is also still room to develop better benchmarks for enzyme discovery, to measure the effectiveness of various models and representations …”

Section: Discovery Of Functional Enzymes With Machine Learningmentioning

confidence: 99%

See 1 more Smart Citation

Opportunities and Challenges for Machine Learning-Assisted Enzyme Engineering

Yang,

Li,

Arnold

2024

ACS Cent. Sci.

View full text Add to dashboard Cite

Enzymes can be engineered at the level of their amino acid sequences to optimize key properties such as expression, stability, substrate range, and catalytic efficiencyor even to unlock new catalytic activities not found in nature. Because the search space of possible proteins is vast, enzyme engineering usually involves discovering an enzyme starting point that has some level of the desired activity followed by directed evolution to improve its “fitness” for a desired application. Recently, machine learning (ML) has emerged as a powerful tool to complement this empirical process. ML models can contribute to (1) starting point discovery by functional annotation of known protein sequences or generating novel protein sequences with desired functions and (2) navigating protein fitness landscapes for fitness optimization by learning mappings between protein sequences and their associated fitness values. In this Outlook, we explain how ML complements enzyme engineering and discuss its future potential to unlock improved engineering outcomes.

show abstract

Section: Annotation Of Enzyme Activity Among Known Proteinsmentioning

confidence: 99%

Section: Discovery Of Functional Enzymes With Machine Learningmentioning

confidence: 99%

Opportunities and Challenges for Machine Learning-Assisted Enzyme Engineering

Yang,

Li,

Arnold

2024

ACS Cent. Sci.

View full text Add to dashboard Cite

show abstract

“…Let 20 be defined as in Section 3.3, where G is the space of probability distributions over the 20 amino acids. f IF (x) is the probability distribution over the mutated site of x given by an inverse folding model.…”

Section: A5 Kernel Proofmentioning

confidence: 99%

Kermut: Composite kernel regression for protein variant effects

Groth,

Kerrn,

Olsen

et al. 2024

Preprint

View full text Add to dashboard Cite

Reliable prediction of protein variant effects is crucial for both protein optimization and for advancing biological understanding. For practical use in protein engineering, it is important that we can also provide reliable uncertainty estimates for our predictions, and while prediction accuracy has seen much progress in recent years, uncertainty metrics are rarely reported. We here provide a Gaussian process regression model, Kermut, with a novel composite kernel for modelling mutation similarity, which obtains state-of-the-art performance for protein variant effect prediction while also offering estimates of uncertainty through its posterior. An analysis of the quality of the uncertainty estimates demonstrates that our model provides meaningful levels of overall calibration, but that instance-specific uncertainty calibration remains more challenging. We hope that this will encourage future work in this promising direction.

show abstract

Fine-tuning protein language models boosts predictions across diverse tasks

Schmirler,

Heinzinger,

Rost

2024

Nat Commun

View full text Add to dashboard Cite

Prediction methods inputting embeddings from protein language models have reached or even surpassed state-of-the-art performance on many protein prediction tasks. In natural language processing fine-tuning large language models has become the de facto standard. In contrast, most protein language model-based protein predictions do not back-propagate to the language model. Here, we compare the fine-tuning of three state-of-the-art models (ESM2, ProtT5, Ankh) on eight different tasks. Two results stand out. Firstly, task-specific supervised fine-tuning almost always improves downstream predictions. Secondly, parameter-efficient fine-tuning can reach similar improvements consuming substantially fewer resources at up to 4.5-fold acceleration of training over fine-tuning full models. Our results suggest to always try fine-tuning, in particular for problems with small datasets, such as for fitness landscape predictions of a single protein. For ease of adaptability, we provide easy-to-use notebooks to fine-tune all models used during this work for per-protein (pooling) and per-residue prediction tasks.

show abstract

FLOP: Tasks for Fitness Landscapes Of Protein wildtypes

Cited by 4 publications

References 55 publications

Opportunities and Challenges for Machine Learning-Assisted Enzyme Engineering

Opportunities and Challenges for Machine Learning-Assisted Enzyme Engineering

Kermut: Composite kernel regression for protein variant effects

Fine-tuning protein language models boosts predictions across diverse tasks

Contact Info

Product

Resources

About