Editorial: Progress and challenges in computational structure-based design and development of biologic drugs
Bibliographic record
Abstract
have emerged as a major class of pharmaceuticals, encompassing monoclonal antibodies, recombinant human proteins and enzymes, fusion proteins, antibody drug conjugates, multi-specific formats, peptides, and vaccines.These modalities serve a wide range of therapeutic areas, including immune-oncology, inflammation, cardiovascular, metabolic, infectious, and rare diseases (DeFrancesco, 2019;Kang and Jung, 2020;Lu et al., 2020;Kaplon et al., 2023).Recent advancements in structure determination, structure prediction, bioanalytical characterization, and machine learning have established in silico approaches as a key toolbox employed in the biologic drug discovery and development pipelines (Fischman and Ofran, 2018;Norman et al., 2020;Fernandez-Quintero et al., 2023).Additionally, physics-based molecular modeling and simulation methods, along with empirical linear models, have matured to routine implementation during biotherapeutic drug candidate selection and optimization.However, the accuracy of these predictions can be improved.Further refinements will be welcomed, particularly towards binding affinity predictions and developability assessments.At the same time, with the fast-paced infusion of artificial intelligence in various research areas that impact daily life, we are witnessing a new chapter being written in biological drug design (Kim et al., 2023).A wide range of nonlinear models, from machine learning to unsupervised deep neural networks and language models, is now emerging.These models, fueled by still modest but expanding biological and structural datasets, are complementing classical methods (Baek et al., 2021;Jumper et al., 2021;Kryshtafovych et al., 2021;Bennett et al., 2023;Lin et al., 2023;Wodak et al., 2023).There are also increasing attempts to integrate advanced computational methods with next-generation sequencing, either from synthetic or natural libraries, to enhance the efficient hit-to-lead optimization and de novo discovery of tight binders with favorable developability profiles (Makowski et al.,
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 | 0.006 | 0.028 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.005 | 0.005 |
| Open science | 0.003 | 0.001 |
| Research integrity | 0.008 | 0.015 |
| Insufficient payload (model declined to judge) | 0.011 | 0.009 |
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".