Stanford's Kundaje Calls Out TRACEBIND Paper for Ignoring ChromBPNet Baseline
anshulkundaje · x · 2026-09-23
Computational biologist Anshul Kundaje posted a four-part critique of the TRACEBIND paper (Tn5 bias correction for ATAC-seq; preprint Oct 2025, published in Nature Communications June 2026):
- Prior work ignored: his group's ChromBPNet code shipped Jan 2023, preprint Dec 2024, and was used in multiple papers before Oct 2025 — yet TRACEBIND neither cites nor compares to it.
- Odd baseline choice: instead it compares to DECIMA, which isn't even trained on chromatin accessibility data.
- Peer review failure: he questions how Nature Communications reviewers missed such an obvious omission.
A notable academic citation dispute touching baseline rigor at a top journal.
More from Research
- CAIS benchmark: GPT-6 Astra automates 20.8% of remote work, up from 2.5% a year ago — scaling01 · 2026-09-23
- Eval scores reported to 5 significant digits despite 2-5 point error bars, researcher flags — dfrsrchtwts · 2026-09-23
- SkillSpec: intent-masked specification reasoning to catch semantic defects in agent skills — Buaa1 · 2026-09-23
- LLMs start overconfident, then swing underconfident when criticized, Nature MI paper finds — ValerioCapraro · 2026-09-23
- Nature MI paper: LLMs start overconfident, then swing underconfident when criticized — ValerioCapraro · 2026-09-23
- UK AISI paper: fixed-budget evals increasingly understate frontier LLM capability — evijit · 2026-09-23