Pith.

103 results for “AlphaFold”

Closest match to your words first, in the title, the authors, the abstract or Pith's review. Among equal matches, newer papers first. The order is not a quality score.

Narrow these results

  1. The paper:

    AlphaFold's Bayesian Roots in Probability Kinematics

    Computer Science · arXiv:2505.19763

    Thomas Hamelryck, Kanti V. Mardia

    First read, from the abstract only

    First read · grok-4.3 · 19 May 2026 · read the abstract only

    Pith has not read this paper's text.

    4 notes from the first read

  2. The paper:

    Calibrating AlphaFold confidence scores for interface quality

    Biology · arXiv:2610.17108

    Sagyndyk Sagadinov, Vsevolod Viliuga, Arne Elofsson

    Pith says

    Machine review · deepseek-v4-flash · 14 Oct 2026 · what was read was not recorded

    Holds up. A clean, bounded application of isotonic calibration to AlphaFold interface confidence scores—useful but incremental.

    1 major objection at §3.2, Table 3 · 5 minor

  4. The paper:

    AF_Cache: Efficient Pipeline for Running AlphaFold for High-Throughput Protein-Protein Interaction Prediction

    Biology · arXiv:2606.04566

    Sarah Narrowe, Arne Elofsson Claudio Mirabello

    Pith says

    Machine review · grok-4.3 · 28 Jun 2026 · what was read was not recorded

    The passes disagreed. Reader: holds up. Skeptic: holds up only with conditions. Referee: major revision.

    AF_Cache packages known optimizations into a Nextflow workflow and shows clear runtime gains on 5050 pairs, but asserts unchanged accuracy without any metrics to back it.

    1 major objection at Abstract

  6. The paper:

    Not Yet AlphaFold for the Mind: Evaluating Centaur as a Synthetic Participant

    Computer Science · arXiv:2508.07887

    Sabrina Namazova, Alessandra Brondetta, Younes Strittmatter and 2 more

    Pith says

    Machine review · deepseek-v4-flash · 5 Aug 2026 · what was read was not recorded

    Holds up, with conditions. Useful first open-loop evaluation of Centaur's generative behavior, with a clear predictive-generative dissociation; the reversal-learning benchmark is synthetic RW data, not human, so the 'divergence from human data' framing overreaches on that task.

    3 major objections, the first at Abstract / 'Reversal Learning Task' in Supplementary … · 4 minor

  7. The paper:

    AlphaFold Database Debiasing for Robust Inverse Folding

    Computer Science · arXiv:2506.08365

    Cheng Tan, Zhenxiao Cao, Zhangyang Gao and 3 more

    Pith says

    Machine review · deepseek-v4-flash · 7 Aug 2026 · what was read was not recorded

    Holds up, with conditions. Solid controlled demonstration that AFDB training degrades inverse folding, but the debiasing claim is only tested on the same paired set—no evidence it works on novel AFDB entries.

    4 major objections, the first at Appendix B.1, C; Sections 6.1–6.3 · 4 minor

  9. The paper:

    Applications of Knot Theory for the Improvement of the AlphaFold Protein Database

    Biology · arXiv:2412.11229

    Pranshu Jahagirdar

    Pith says

    Machine review · deepseek-v4-flash · 11 Aug 2026 · what was read was not recorded

    Does not hold up as stated. A clear, earnest short paper whose Alexander-Briggs comparison is fine but whose headline Gauss-code mismatch claim is invalid because the comparison ignores that Gauss codes are representation-dependent.

    This review says the paper's mathematics is wrong. Pith did not check that against the paper's source.

    4 major objections, the first at Data Collection and Analysis; Table 1 · 5 minor

  10. The paper:

    Rapid response to fast viral evolution using AlphaFold 3-assisted topological deep learning

    Biology · arXiv:2411.12370

    JunJie Wee, Guo-Wei Wei

    Pith says

    Machine review · deepseek-v4-flash · 12 Aug 2026 · what was read was not recorded

    Holds up, with conditions. AF3 structures work nearly as well as experimental ones inside MT-TopLap, but the rapid-response claim overreaches: the HK.3 test fine-tunes on the variant's own DMS before cross-validating.

    4 major objections, the first at Section 2 (HK.3 validation) · 6 minor

  14. The paper:

    AlphaFold Meets Flow Matching for Generating Protein Ensembles

    Biology · arXiv:2402.04845

    Bowen Jing, Bonnie Berger, Tommi Jaakkola

    Pith says

    Machine review · deepseek-v4.1-flash · 11 Nov 2026 · what was read was not recorded

    Holds up, with conditions. A well-baselined, code-backed method paper whose MD generalization claim is weaker than its framing because the ATLAS split is temporal, not structural.

    5 major objections, the first at §4.3, abstract:s4 · 8 minor