{"doi":"10.7302/3896","title":"Predictions of Antibody Biophysical Properties for Improving Drug Development","abstract":"Antibodies are natural proteins that are central to the function of the immune system. With high affinity and specificity, antibodies serve as adaptor molecules between recognition of foreign invaders and recruitment of immune cells. Monoclonal antibodies (mAbs) are attractive as therapeutic because of their drug-like properties (high solubility, stability, and specificity as well as excellent pharmacokinetics). However, the process of discovering and developing antibody drugs is time-consuming and expensive. Majority of the lead molecules fail in the clinical trial and only a few of them can survive from the clinical trial and finally be considered as drug. As a result, identifying antibodies with drug-like properties at early stages of the discovery process can reduce the risk of their rejection at late stages of development as well as reduce the time and cost of drug development. Our goal is to develop physicochemical descriptors and machine learning models for identifying and improving antibodies with drug-like properties during early stages of clinical development based on sequence based and predicted structure-based features. We argued that antibody specificity is the most important property to identify drug-like molecules since the measurements for self-interaction and non-specific interaction displayed a strong statistical significance in segregating approved drugs and those in clinical trial Phase II and III. The identified descriptors were used to guide the antibody library design to further improve antibody specificity while maintaining or increasing the humanness. Chapter 1 describes the reason why antibodies are considered as potential class of therapeutics and the background and motivation of molecular assessment for drug development; Chapter 2 shows how we developed the chemical rules for identifying monoclonal antibodies with high specificity defined by previously reported experimental measurements; Chapter 3 shows a discovery that antibody variants with mutations in complementarity-determining regions (CDRs) that reduce non-specific binding also display drug-like specificity using a large set of human library antibodies; Chapter 5 describes the summary of the findings and discusses future directions in this field.","journal":"Deep Blue (University of Michigan)","year":2021,"id":229831,"datarank":0.0,"base_score":0.0,"endowment":0.0,"self_citation_contribution":0.0,"citation_network_contribution":0.0,"self_endowment_contribution":0.0,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":0,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.9524,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2021-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":690845,"name":"Yulei Zhang","orcid":"0000-0002-4307-9850","position":0,"is_corresponding":true}],"reference_count":0,"raw_metadata":null,"created_at":"2026-07-18T23:55:01.675482Z","pmid":null,"pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}