When chromatin models disagree: null accessibility shift but a disrupted CEBPG binding site
jmschreiber91 · x · 2026-10-31
Follow-up from jmschreiber91: an interesting case where the models "disagree" — a variant shows essentially no effect on chromatin accessibility (DNase-seq lfc ≈ 0) yet the CEBPG TF binding model infers a strongly disrupted binding site (delta -1.8). Possible explanations: the TF may not be a pioneer factor, or one of the models may simply be wrong. Track images show the DNase-seq and CEBPG footprints around the variant in K562 cells.
More from Research
- New MolmoSpaces-CommonSense benchmark: no robot policy tops 30% on manipulation tasks — RanjayKrishna · 2026-10-31
- Mischief launches video model trained from lab-grown brain cells at 0.5¢ per second — evilsocket · 2026-10-31
- Stanford CS336 notes: Percy Liang on scaling recipes and hyperparameter transfer — stanfordnlp · 2026-10-31
- Stanford paper proposes rewriting scaling law: optimal synthetic data share rises to 32% with compute — stanfordnlp · 2026-10-31
- Composable LoRA directions for texture, geometry and lighting make 3D fine-tuning cheaper — burkov · 2026-10-31
- Moondream releases RefCOCO-M with pixel-accurate masks and cleaned prompts for grounding evals — victormustar · 2026-10-31