Newsroom2026-01-10 → 2026-04-10 Dispatches from the claim feed
Every typed claim posted to Polymer, in chronological order. Each dispatch carries its outcome, method, scale, and topology. For the structural view of the same corpus, open the claims universe.
10Dispatches
2026-01-10 → 2026-04-10Timeline
v2 · 21 featuresSchema
APR 10 · 2026Exp 22Well-defended rule CpG Erosion Chronometer
POSITIVE
CpG erosion tracks TE evolutionary age — 12 of 13 biophysical parameters correlate with TE age. SINEs maintain a reservoir (0.86 obs/exp young); universal erosion floor 0.15. Roll exception confirms the rule via arrangement-capacity buffering.
Effect 0.283·Partial corr·1.0 kb·4.6M instances·3 layers·transposome
APR 07 · 2026Exp 21Qualified / proof Arrangement Envelope (LP Proof)
POSITIVE
Biophysical arrangement capacity varies 9–87% across 13 parameters at FIXED GC content. LP optimization establishes bounds; this is a proof, not a statistical finding.
Effect 0.272·Proof·3 bp·65 instances·1 layers·biophysics
MAR 30 · 2026Exp 19Qualified / proof Rapid-Response Gene Biophysics
QUALIFIED
Rapid-response genes (immediate-early, stress) show distinct flanking biophysics vs housekeeping. Signal survives GC control but is weaker than gene-body signal from Exp 18.
Effect 0.050·Partial corr·10 kb·501 instances·3 layers·gene_expression
MAR 25 · 2026Exp 18Well-defended rule Gene-Flanking Biophysics
POSITIVE
DNA deformability predicts gene expression in gene bodies, GC-independent. Partial r=−0.345 (body); curvature +0.285. 5 of 6 hypotheses supported; promoter deformability failed.
Effect 0.129·Partial corr + Enrichment·10 kb·20.0K instances·5 layers·gene_expression
MAR 18 · 2026Exp 17Simple positive Crossover Hotspot Biophysics
STRONG POSITIVE
Meiotic crossover hotspots are biophysically distinct from non-crossover sites. AUROC 0.827 without GC. CO and NCO are also distinct from each other.
Effect 0.262·ML·1.0 kb·89.1K instances·4 layers·recombination
MAR 10 · 2026Exp 16Simple positive HLA Expression-Biophysics
POSITIVE
HLA-A allele expression correlates with non-coding biophysics. HLA-A dG37 vs TPM r=−0.76; within-protein Cohen's d=2.49; HLA-A > HLA-B > HLA-C gradient tracks correlation strength.
Effect 0.246·Statistical·50 kb·100 instances·3 layers·hla
MAR 02 · 2026Exp 15Simple positive TE-Gene Correlation Atlas
POSITIVE
TE class correlates with gene expression, GC-independent: SINEs +0.18, LTRs −0.18, LINEs −0.14 (partial r). Random forest R²=37.7%.
Effect 0.072·ML + Partial corr·10 kb·20.0K instances·3 layers·transposome
FEB 14 · 2026Exp 11Null / failed CDI Negative
NEGATIVE
Q-weighted entropy (CDI) is not distinct from ThermAge on 450K arrays — r=0.9991 between them.
Effect 0.300·Statistical·447 bp·446.7K instances·2 layers·clocks
JAN 28 · 2026Exp 09Null / failed Nucleosome Positioning FAIL
FAIL
Physics does not predict MNase nucleosome positioning at 1 kb resolution. Periodicity r=−0.03; after GC control, partial r=−0.009 (signal is GC).
Effect 0.013·Partial corr·1.0 kb·3.0M instances·3 layers·chromatin
JAN 10 · 2026Exp 03Well-defended rule TE Silencing Mechanism
POSITIVE
DNA biophysical properties predict TE silencing mechanism (H3K9me3 vs H3K27me3). Stacking energy is the top feature; signal survives GC and within-age controls.
Effect 0.235·ML·148 bp·4.6M instances·4 layers·transposome
— End of feed —