
ALT Editorial BioAI Frontier cover for the Anthropic × BigHat preprint “Reasoning in the loop.” Text reads Anthropic × BigHat; Claude Helps Turn Hypotheses into Experiments; Antibody Lead Optimization; and 7 Rounds · ~1,010 Variants · Preprint, over an abstract antibody-and-measurement illustration on a dark field. Claude is emphasized in yellow. This is an original editorial generated cover, not an official Anthropic or BigHat visual and not a figure from the paper.

ALT Figure 4 from the paper. Panel (a) shows a binding-assay curve for the VHH (nanobody) parent with a 1:1 kinetic fit. Panel (b) shows the same parent’s thermal assay: fluorescence and 266 nm scattered light versus temperature, with model fits. Scattering rises while fluorescence is still almost unchanged, illustrating aggregation beginning while the protein is still folded, so a reported melting or aggregation temperature can mislead. Unaltered official figure from Stanton, Solomon, Delaney, Montgomery, Humphries, Acker, Henrici, Eid, Abrams and Greenside, “Reasoning in the loop: antibody lead optimization by probing mechanism and measurement,” arXiv:2611.19812, https://arxiv.org/abs/2611.19812, licensed under CC BY 4.0, https://creativecommons.org/licenses/by/4.0/.

ALT Figure 5 from the paper. Averaged thermal-assay traces for parents and leads: fluorescence (top) and 266 nm scattered light (bottom) versus temperature for the VHH/nanobody (a) and the scFv (b). The VHH lead shows little scattering; the scFv lead remains in solution to a higher temperature than its parent. Additional panels test the physical model by changing protein concentration and heating rate against preregistered predictions. Unaltered official figure from Stanton, Solomon, Delaney, Montgomery, Humphries, Acker, Henrici, Eid, Abrams and Greenside, “Reasoning in the loop: antibody lead optimization by probing mechanism and measurement,” arXiv:2611.19812, https://arxiv.org/abs/2611.19812, licensed under CC BY 4.0, https://creativecommons.org/licenses/by/4.0/.

ALT Figure 6 from the paper. Final-round designs plotted against the revised specifications. (a) VHH/nanobody off-rate versus the aggregation metric p k_agg. (b) scFv off-rate versus the solution-collapse temperature T_collapse, the temperature at which the sample falls out of solution. Dashed lines mark constraints; the selected leads are indicated. Unaltered official figure from Stanton, Solomon, Delaney, Montgomery, Humphries, Acker, Henrici, Eid, Abrams and Greenside, “Reasoning in the loop: antibody lead optimization by probing mechanism and measurement,” arXiv:2611.19812, https://arxiv.org/abs/2611.19812, licensed under CC BY 4.0, https://creativecommons.org/licenses/by/4.0/.

ALT Editorial BioAI Frontier cover for Isomorphic Labs’ $5 billion Series C. Text reads ISOMORPHIC LABS, $5 BILLION, SERIES C, and Scaling AI Drug Discovery, over an abstract protein-and-ligand illustration on a dark field. This is an original editorial generated cover, not an official Isomorphic Labs or Alphabet visual.

ALT BioAI Frontier original cover for an external analysis of Claude Fable 5.2. Headline: Claude Fable 5.2. Hook: Outperforms Protein Language Models. Qualifier: In next-amino-acid prediction. Footer: External analysis · Tested open models. Abstract teal molecular ribbons on a deep navy field. Typography and background are an original BioAI Frontier cover; the background was first-party generated for this post and is not a figure from Xinming Tu and Leo Tianlai Chen's analysis or the LLM-bioseq-probe repository.

ALT Chart from Xinming Tu and Leo Tianlai Chen's 16 November 2026 analysis: next-amino-acid prediction accuracy of Claude models compared with open protein language models, including ProGen3, ProtGPT3 and ESM, on real protein sequences versus shuffled-sequence controls. Copyright (c) 2026 Xinming Tu. Available under the MIT License: permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, subject to the following condition: the above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. Full license: https://github.com/XinmingTu/LLM-bioseq-probe/blob/main/LICENSE

ALT Chart from Xinming Tu and Leo Tianlai Chen's 16 November 2026 analysis: next-amino-acid prediction results for 22 general-purpose language models, including Claude, GPT, Gemini and open-weight models, comparing accuracy on real sequences with shuffled-sequence controls. Copyright (c) 2026 Xinming Tu. Available under the MIT License: permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, subject to the following condition: the above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. Full license: https://github.com/XinmingTu/LLM-bioseq-probe/blob/main/LICENSE

ALT Chart from Xinming Tu and Leo Tianlai Chen's 16 November 2026 analysis: how each model's real-versus-shuffled next-amino-acid prediction advantage changes with protein-family size, a pattern also seen in dedicated protein language models. Copyright (c) 2026 Xinming Tu. Available under the MIT License: permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, subject to the following condition: the above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. Full license: https://github.com/XinmingTu/LLM-bioseq-probe/blob/main/LICENSE

ALT BioAI Frontier editorial cover for OpenScientist in Nature Medicine. Text highlights a six-blood-protein survival-risk score, c-index 0.826, tested on 176 unseen participants. This is one of five biomedical case studies; the hidden cohort included 17 deaths. Six conceptual beads connect to a score tile on a navy background. The illustration is decorative, not molecular or experimental data.

ALT Original Figure 1: OpenScientist investigation workflow. A scientist specifies a task and optionally provides data. The agent searches PubMed, reads papers, writes and runs code, and analyzes files using scientific skills. Findings, hypotheses, plots and transcripts enter a persistent knowledge state; optional user feedback guides later iterations. A final report is compiled. This is computational analysis, not an automated wet-lab experiment. Source: Roberts et al., Nature Medicine (2026), Fig. 1, CC BY 4.0; rendered from accepted manuscript.

ALT Original Figure 4: plasma proteomic survival modeling. Panel a shows demographics and plasma proteomics from 500 participants supplied with a survival-modeling task. Panel b compares concordance indices; panel c compares with biological-clock models. The dagger note identifies AI Agent (OwnSplit) as internal 50/50 train-test validation, not the competition holdout. These evaluation settings must not be conflated with the separate 176-person hidden-cohort rerun reported in the text. Original axes and manuscript markings retained. Source: Roberts et al., Nature Medicine (2026), Fig. 4, CC BY 4.0.

ALT Original Figure 3: Alzheimer's biomarker analysis. Panel a shows the SEABIRD proposal, prespecified analysis plan and data from 325 participants. Panel b compares human ROC curves on the left with OpenScientist curves on the right, retaining biomarker legends, AUCs and confidence intervals. Both identify pT217/T217 as the strongest PET-status discriminator. This is the final optimized run after prompt refinement and data curation. Original manuscript markings retained. Source: Roberts et al., Nature Medicine (2026), Fig. 3, CC BY 4.0.

ALT BioAI Frontier editorial cover with the headline Baker Lab × Green Lab and text Choose the C–H Site. Design the Enzyme. Supporting text states One Molecule. Two Target Sites. Estradiol: C6 or C16. Preprint, wet-lab validated. A stylized folded-protein pocket holding an amber substrate represents the design concept; this is an illustration, not a measured molecular structure.

ALT Original Figure 4: site-selective estradiol hydroxylation. A shows C6 hydroxylation to 6alpha-hydroxyestradiol or C16 hydroxylation to estriol. B shows 96 C6 designs and 67 C16 designs; initial yields were below 1% and about 1%, with 93% and >99% diastereomeric excess. C shows Q20N, W70K, I121V and V140Q substitutions increasing C16 yield to 37% and turnovers to 37 while retaining >99% d.e. D overlays the substrate-free, heme-bound dOHes-2-NKVQ crystal structure with the design, C-alpha RMSD 0.65 angstrom. These percentages for d.e. describe stereoselectivity, not yield. Source: Kalvet et al., bioRxiv (2026), Fig. 4, CC BY 4.0.

ALT Original Figure 1: design principles and workflow for de novo C-H hydroxylases. A-D show hydroxylation applications, oxygen/peroxide activation, the Glu-His dyad promoting heterolytic peroxide cleavage, and Compound I hydrogen abstraction and radical rebound. E-F show the minimal heme-based catalytic motif and target substrate geometry, RFdiffusion2/3 backbone generation, iterative LigandMPNN/Rosetta Relax sequence design, and Chai-1 or AlphaFold3 filtering. Original panels and labels retained. Source: Kalvet et al., bioRxiv (2026), Fig. 1, CC BY 4.0.

ALT Fig. 1. MDClaw uses an execution graph to constrain tool calls, keep auditable records, and support branching workflows. a, A coding agent uses the harness to plan and run MD for a human or AI scientist’s question. b, 100 command-line tools and 14 domain skills cover preparation, simulation and analysis. c, DAG rules reject invalid stage transitions, inputs and settings, and let results be audited. d, Skills guide planning; tools execute the plan and support branching, such as sequence variants, force-field changes and replicas.

ALT Fig. 2. MDDataBench task systems, prompts and artifact-based scoring. a, 31 test systems, shown from deposited structures, are held out from 128 development tasks. b, An example 2PQL prompt with required artifacts. c, Submissions are scored on preparation and MD checks against reference data, not on the agent’s written report. d, The same prompt is run under three conditions; the primary metric is pass^3, the share of systems that pass all three repeats.













































