Revealing the hidden sequence distribution of epitope-specific TCR repertoires and its influence on machine learning model performance DOI Creative Commons
Sofie Gielis, Maria Chernigovskaya, Milena Pavlović

и другие.

Опубликована: Окт. 24, 2024

Abstract Numerous efforts have been made to decipher the epitope-T cell receptor (TCR) recognition code. Both simple machine learning techniques and deep strategies used train models predict binding of epitopes by TCR sequences. A good training data set rests at basis every accurate prediction model, yet little attention has given composition these sets. In this paper, we studied natural distribution sequences within epitope-specific repertoires, i.e. a TCRs same epitope, its impact on predictability TCR-epitope interactions. We found that observed diversity repertoires can result from smaller core motifs constrained generation. Moreover, clear relationship was between sequence performance metrics, emphasizing importance ground-truth when using in domain. Taken together, findings inform help push epitope-TCR next level.

Язык: Английский

Contrastive learning of T cell receptor representations DOI Creative Commons
Yuta Nagano,

Andrew G. T. Pyo,

Martina Milighetti

и другие.

Cell Systems, Год журнала: 2025, Номер unknown, С. 101165 - 101165

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

0
The Evolving T Cell Receptor Recognition Code: The Rules Are More Like Guidelines DOI Open Access

George I. Gray,

P. Chukwunalu Chukwuma,

Bassant Eldaly

и другие.

Immunological Reviews, Год журнала: 2025, Номер 329(1)

Опубликована: Янв. 1, 2025

ABSTRACT αβ T cell receptor (TCR) recognition of peptide–MHC complexes lies at the core adaptive immunity, balancing specificity and cross‐reactivity to facilitate effective antigen discrimination. Early structural studies established basic frameworks helpful for understanding contextualizing TCR features such as peptide MHC restriction. However, growing database launched from work continue reveal exceptions common assumptions simplifications derived earlier work. Here we explore our evolving recognition, illustrating how biophysical investigations regularly uncover complex phenomena that push against paradigms expand TCRs bind discriminate between peptide/MHC complexes. We discuss implications these findings basic, translational, predictive immunology, including challenges in accounting inherent adaptability, flexibility, occasional sloppiness characterize recognition.

Язык: Английский

Процитировано

0
Reading the repertoire: Progress in adaptive immune receptor analysis using machine learning DOI Creative Commons

Timothy J O'Donnell,

Chakravarthi Kanduri, Giulio Isacchini

и другие.

Cell Systems, Год журнала: 2024, Номер 15(12), С. 1168 - 1189

Опубликована: Дек. 1, 2024

Язык: Английский

Процитировано

2
Revealing the hidden sequence distribution of epitope-specific TCR repertoires and its influence on machine learning model performance DOI Creative Commons
Sofie Gielis, Maria Chernigovskaya, Milena Pavlović

и другие.

Опубликована: Окт. 24, 2024

Abstract Numerous efforts have been made to decipher the epitope-T cell receptor (TCR) recognition code. Both simple machine learning techniques and deep strategies used train models predict binding of epitopes by TCR sequences. A good training data set rests at basis every accurate prediction model, yet little attention has given composition these sets. In this paper, we studied natural distribution sequences within epitope-specific repertoires, i.e. a TCRs same epitope, its impact on predictability TCR-epitope interactions. We found that observed diversity repertoires can result from smaller core motifs constrained generation. Moreover, clear relationship was between sequence performance metrics, emphasizing importance ground-truth when using in domain. Taken together, findings inform help push epitope-TCR next level.

Язык: Английский

Процитировано

1