Quantification of Uncertainty in Peptide-MHC Binding Prediction Improves High-Affinity Peptide Selection for Therapeutic Design

Zeng, Haoyang; Gifford, David K.

doi:10.1016/j.cels.2019.05.004

Cited by 59 publications

(74 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this work we will study the affinity of molecules (Lan et al, 2019) to the receptor sites to confirm that all molecules are attached to the same receptor site (Zeng and Gifford, 2019), a comparative study of the bonds between each molecule with the receptor site helps us to know the molecule that has a great affinity so a better effect (Aviñó et al, 2019).…”

Section: Methods and Computational Detailsmentioning

confidence: 99%

The Development of New Molecules Having Antiemetic Activity Using Molecular Modeling

Kanouni¹,

Benguerba²

2019

Current Research in Bioinformatics

View full text Add to dashboard Cite

The solubility of a drug in water or in the blood represents the most desired parameter in medicine. Our aim is to obtain molecules with properties more effective than those of metoclopramide. In this work, two molecules with high solubility are constructed. Metoclopramide (a benzamide derivative) is a dopamine receptor antagonist used as an antiemetic drug. Its solubility in water is 200 mg/L at 25°C. In this work, we will develop two other molecules that have the same therapeutic activity of metoclopramide with a higher solubility in water, therefore in the blood, without affecting the other properties. The two molecules developed by molecular modeling with a chemical modification of the OH group of metoclopramide have a high solubility: approximately 3 times and 8 times that of metoclopramide. For the other physicochemical properties, there is a great similarity between the molecules. Thus, the two proposed molecules will have antiemetic activity, the second molecule will be more favorable because of its higher solubility and the number of HBA and HBD.

show abstract

Section: Methods and Computational Detailsmentioning

confidence: 99%

The Development of New Molecules Having Antiemetic Activity Using Molecular Modeling

Kanouni¹,

Benguerba²

2019

Current Research in Bioinformatics

View full text Add to dashboard Cite

show abstract

“…For a given MHC class II molecule, we train a neural network-based machine learning (ML) model (PUFFIN) [10] that takes a 9 residue peptide sequence as input, and outputs a measure of the strength of the peptide-MHC interaction. PUFFIN outputs uncertainty estimates which allows us to compute Bayesian acquisition functions.…”

Section: We Evaluate the Complete Anchor Substitution Landscape With mentioning

confidence: 99%

“…We trained a neural network-based machine learning (ML) model (PUFFIN) [10] to predict the enrichment label of a new peptide. The predictor takes a 9 residue peptide sequence as input, and outputs an enrichment label.…”

Section: Training a Neural Network Based ML Model To Predict The Enrimentioning

confidence: 99%

“…PUFFIN uses an ensemble of deep residual networks that takes as input the peptide and MHC amino acid sequences and outputs a predicted affinity distribution of the peptide for the MHC, achieving state of the art performance on class II MHC binding prediction tasks [10]. We show that our method generates peptide modifications that improve peptide binding affinity for two class II MHCs.…”

Section: Introductionmentioning

confidence: 99%

“…For our objective function, we use predictions from the PUFFIN peptide-MHC binding model [10] trained on peptide binding data from a class II MHC yeast display platform [11].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Machine learning optimization of peptides for presentation by class II MHCs

Dai¹,

Huisman

Zeng³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

T cells play a critical role in normal immune responses to pathogens and cancer and can be targeted to MHC-presented antigens via interventions such as peptide vaccines. Here, we present a machine learning method to optimize the presentation of peptides by class II MHCs by modifying the peptide's anchor residues. Our method first learns a model of peptide affinity for a class II MHC using an ensemble of deep residual networks, and then uses the model to propose anchor residue changes to improve peptide affinity. We use a high throughput yeast display assay to show that anchor residue optimization successfully improved peptide binding.

show abstract

Deep Learning in Proteomics

et al. 2020

View full text Add to dashboard Cite

Proteomics, the study of all the proteins in biological systems, is becoming a data-rich science. Protein sequences and structures are comprehensively catalogued in online databases. With recent advancements in tandem mass spectrometry (MS) technology, protein expression and post-translational modifications (PTMs) can be studied in a variety of biological systems at the global scale. Sophisticated computational algorithms are needed to translate the vast amount of data into novel biological insights. Deep learning automatically extracts data representations at high levels of abstraction from data, and it thrives in data-rich scientific research domains. Here, a comprehensive overview of deep learning applications in proteomics, including retention time prediction, MS/MS spectrum prediction, de novo peptide sequencing, PTM prediction, major histocompatibility complex-peptide binding prediction, and protein structure prediction, is provided. Limitations and the future directions of deep learning in proteomics are also discussed. This review will provide readers an overview of deep learning and how it can be used to analyze proteomics data.

show abstract

Quantification of Uncertainty in Peptide-MHC Binding Prediction Improves High-Affinity Peptide Selection for Therapeutic Design

Cited by 59 publications

References 27 publications

The Development of New Molecules Having Antiemetic Activity Using Molecular Modeling

The Development of New Molecules Having Antiemetic Activity Using Molecular Modeling

Machine learning optimization of peptides for presentation by class II MHCs

Deep Learning in Proteomics

Contact Info

Product

Resources

About