Computational scanning for responding clonotypes in immune repertoires

Introduction

Personalized medicine promises to exploit the diversity of human immune repertoires to design targeted T cell and B cell–based vaccines and therapeutics. However, even for mass vaccine design, identifying responding clonotypes to new viruses is costly, labor-intensive, and time-consuming.

Project Overview

RESPOND turns existing algorithmic solutions for identifying responding immune clonotypes into a user-friendly platform to aid drug and therapeutics discovery. RESPOND combines different algorithmic solutions and finds the best computational approach to fit the user’s needs.

Methodology

It is based on four methods developed during the ERC CoG STRUGGLE project:

1. ALICE
2. NoiSET
3. fSTAR
4. HLA-Guessr

These methods are based on different properties of responding clonotypes, from sequence similarity to abundance and publicness. The idea of RESPOND is to integrate these features in a user-friendly tool that exploits the strengths of all methods to find the best answer to the practitioner’s query.

Output

RESPOND will return lists of candidates for responding clonotypes and a statistical analysis of their occurrence in databases.

Collaboration

Collaborating with big pharma, start-ups, medical researchers, and practitioners, RESPOND will develop computational solutions to reduce the costs of biotechnological discovery and significantly decrease the time to test new vaccines and treatments.

User Experience

Based on consulting and feedback, we will aim to give the user what they need in an appealing interface. The tool will be of use for everyone involved in exploiting immune repertoires for treatment and prophylactics, from personalized medicine and mass vaccine design to agriculture.

Commercialization

RESPOND will pursue commercialization and market solutions both within academia and the biotechnology and medical sectors.