Stability-Diversity Tradeoffs Impose Fundamental Constraints on Selection of Synthetic Human VH/VL Single-Domain Antibodies from In Vitro Display Libraries
Bibliographic record
Abstract
Human autonomous VH/VL single-domain antibodies (sdAbs) are attractive therapeutic molecules, but often suffer from suboptimal stability, solubility and affinity for cognate antigens. Most commonly, human sdAbs have been isolated from in vitro display libraries constructed via synthetic randomization of rearranged VH/VL domains. Here, we describe the design and characterization of three novel human VH/VL sdAb libraries through a process of: (i) exhaustive biophysical characterization of 20 potential VH/VL sdAb library scaffolds, including assessment of expression yield, aggregation resistance, thermostability and tolerance to CDR substitutions; (ii) in vitro randomization of the CDRs of three VH/VL sdAb scaffolds, with tailored amino acid representation designed to promote solubility and expressibility; and (iii) systematic benchmarking of the three VH/VL libraries by panning against five model antigens. We isolated ≥1 antigen-specific human sdAb against four of five targets (13 VHs and 7 VLs in total); these were predominantly monomeric, had antigen-binding affinities ranging from 5 nM – 12 µM (average: 2 – 3 µM), but had highly variable expression yields (range: 0.1 – 19 mg/L). Despite our efforts to identify the most stable VH/VL scaffolds, selection of antigen-specific binders from these libraries was unpredictable (overall success rate for all library-target screens: ~53%) with a high attrition rate of sdAbs exhibiting false positive binding by ELISA. By analyzing VH/VL sdAb library sequence composition following selection for monomeric antibody expression (binding to protein A/L followed by amplification in bacterial cells), we found that some VH/VL sdAbs had marked growth advantages over others, and that the amino acid composition of the CDRs of this set of sdAbs was dramatically restricted (bias towards Asp and His and away from aromatic and hydrophobic residues). Thus, CDR sequence clearly dramatically impacts the stability of human autonomous VH/VL immunoglobulin domain folds, and sequence-stability tradeoffs must be taken into account during the design of such libraries.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".