Data-driven engineering of protein therapeutics

• Published in: Current opinion in biotechnology Vol. 60; pp. 104 - 110
• Main Authors: Faber, Matthew S, Whitehead, Timothy A
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.12.2019
• Subjects: Antibodies; Humans; Protein Engineering
• ISSN: 0958-1669; 1879-0429
• DOI: 10.1016/j.copbio.2019.01.015

[Display omitted] •Low-cost reading and writing of DNA expands protein sequence-function datasets.•Analysis of natural sequences informs mutant selection for satisfying constraints.•B-cell antibody repertoires provide starting sequences and optimal VH–VL pairings.•Combination and integration of datasets accelerates therapeutic protein development. Protein therapeutics requires a series of properties beyond biochemical activity, including serum stability, low immunogenicity, and manufacturability. Mutations that improve one property often decrease one or more of the other essential requirements for therapeutic efficacy, making the protein engineering challenge difficult. The past decade has seen an explosion of new techniques centered around cheaply reading and writing DNA. This review highlights the recent use of such high throughput technologies for engineering protein therapeutics. Examples include the use of human antibody repertoire sequence data to pair antibody heavy and light chains, comprehensive mutational analysis for engineering antibody specificity, and the use of ancestral and inter-species sequence data to engineer simultaneous improvements in enzyme catalytic efficiency and stability. We conclude with a perspective on further ways to integrate mature protein engineering pipelines with the exponential increases in the volume of sequencing data expected in the forthcoming decade.