Machine Learning Designs Non-Hemolytic Antimicrobial Peptides

Capecchi, Alice; Cai, Xingguang; Personne, Hippolyte; Köhler, Thilo; Delden, Christian van; Reymond, Jean‐Louis

doi:10.26434/chemrxiv.14233418

Cited by 6 publications

(26 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This system has taken into account a powerful tool to learn from data (Lee et al., 2018; Li et al., 2018). It allows users to generate new molecules and predict their characteristics and activity using the experimental structure of known drugs (Capecchi et al., 2021). Now, the machine learning model has been applied to the prediction of bioactive peptides.…”

Section: Introductionmentioning

confidence: 99%

Research progress in the screening and evaluation of umami peptides

Gao

Pan

et al. 2022

Comp Rev Food Sci Food Safe

View full text Add to dashboard Cite

Umami is an important element affecting food taste, and the development of umami peptides is a topic of interest in food‐flavoring research. The existing technology used for traditional screening of umami peptides is time‐consuming and labor‐intensive, making it difficult to meet the requirements of high‐throughput screening, which limits the rapid development of umami peptides. The difficulty in performing a standard measurement of umami intensity is another problem that restricts the development of umami peptides. The existing methods are not sensitive and specific, making it difficult to achieve a standard evaluation of umami taste. This review summarizes the umami receptors and umami peptides, focusing on the problems restricting the development of umami peptides, high‐throughput screening, and establishment of evaluation standards. The rapid screening of umami peptides was realized based on molecular docking technology and a machine learning method, and the standard evaluation of umami could be realized with a bionic taste sensor. The progress of rapid screening and evaluation methods significantly promotes the study of umami peptides and increases its application in the seasoning industry.

show abstract

Section: Introductionmentioning

confidence: 99%

Research progress in the screening and evaluation of umami peptides

Gao

Pan

et al. 2022

Comp Rev Food Sci Food Safe

View full text Add to dashboard Cite

show abstract

“…Our predictive model for the solubility task has the lowest accuracy of 69.0% amongst all, and this is mostly attributed to the difficulty associated with solubility prediction in bioinformatics. The one-hot representation of peptides followed by an RNN results the best hemolysis model in terms of AUROC in [64]. The choice of one-hots requires training features specific to each position though, so we do not expect the model to generalize.…”

Section: Resultsmentioning

confidence: 99%

Serverless Prediction of Peptide Properties with Recurrent Neural Networks

Ansari

White

2022

Preprint

View full text Add to dashboard Cite

We present three deep learning sequence prediction models for hemolysis, solubility, and resistance to non-specific interactions of peptides that achieve comparable results to the state-of-the art models. These predictive models share a common architecture of bidirectional recurrent neural networks (LSTM). These models are implemented in JavaScript so that they can be run on a static website without use of a dedicated server. This removes the cost, and long-term management of a server, while still enabling open and free access to the models. This "serverless" prediction model is a demonstration of edge computing bioinformatics and removes the dependence on cloud providers or self-hosting of resource-rich academic institutions. This is feasible because of the continued track of Moore's law and ubiquitous hardware acceleration of deep learning computations on new phones and desktops.

show abstract

“…Moreover, XGBC-Hem performs feature selection to improve the performance of the model. Lastly, the dataset from [12], referred to in this work as RNN-Hem, is selected as the third benchmark, since this work uses a sequence-based deep learning model to predict the hemolytic activity of peptides. All three datasets offer training sets and independent test sets consisting of therapeutic peptides and antimicrobial peptides, which will be used to train and evaluate the classification pipeline.…”

Section: Data Aggregationmentioning

confidence: 99%

AMPDeep: Hemolytic Activity Prediction of Antimicrobial Peptides using Transfer Learning

Salem

Arshadi

Yuan

2022

Preprint

View full text Add to dashboard Cite

Background: Deep learning’s automatic feature extraction has proven to give superior performance in many sequence classification tasks. However, deep learning models generally require a massive amount of data to train, which in the case of Hemolytic Activity Prediction of Antimicrobial Peptides creates a challenge due to the small amount of available data. Results: Three different benchmarks for hemolysis activity prediction of therapeutic and antimicrobial peptides are gathered and the AMPDeep pipeline is implemented for each. The result demonstrate that AMPDeep outperforms the previous works on all three benchmarks, including works that use physicochemical features to represent the peptides or those who solely rely on the sequence and use deep learning to learn representation for the peptides. AMPDeep fine-tunes a large transformer based model on a small amount of peptides and successfully leverages the patterns learned from other protein and peptide databases to assist hemolysis activity prediction modeling. Conclusions: In this work transfer learning is leveraged to overcome the challenge of small data and a deep learning based model is successfully adopted for hemolysis activity classification of antimicrobial peptides. This model is first initialized as a protein language model which is pre-trained on masked amino acid prediction on many unlabeled protein sequences in a self-supervised manner. Having done so, the model is fine-tuned on an aggregated dataset of labeled peptides in a supervised manner to predict secretion. Through transfer learning, hyper-parameter optimization and selective fine-tuning, AMPDeep is able to achieve state-of-the-art performance on three hemolysis benchmarks using only the sequence of the peptides. This work assists the adoption of large sequence-based models for peptide classification and modeling tasks in a practical manner.

show abstract

Machine Learning Designs Non-Hemolytic Antimicrobial Peptides

Cited by 6 publications

References 0 publications

Research progress in the screening and evaluation of umami peptides

Research progress in the screening and evaluation of umami peptides

Serverless Prediction of Peptide Properties with Recurrent Neural Networks

AMPDeep: Hemolytic Activity Prediction of Antimicrobial Peptides using Transfer Learning

Contact Info

Product

Resources

About