The deposited dm6 spike-in table covers the eighteen H3K79me2 samples and nothing else - the ten MEN1 and DOT1L ChIPs in the same series have no fly chromatin, so every occupancy number from them is a within-sample contrast and no gain-or-loss statement on an absolute scale is available
also called GSE149183 factor-ChIP scale gap
This study included a clever internal ruler that lets you say how much of a signal was really lost. The ruler was only put into one half of the experiments, and not into the half this analysis needed it for.
The same thing, for a scientist
The series processing line reads *'The **H3K79me2 samples** also contain Drosophila spike-in chromatin as described; **these samples** were additionally aligned to dm6'*, and the two deposited supplementary tables are named for H3K79me2 [E1]. Both were opened: each carries a five-column spike-in block (`Sample Names`, `hg19/mm9 PF_READS_ALIGNED`, `dm6 PF_READS_ALIGNED`, `% Fly reads/total reads`, `Ratio to DMSO control`) with exactly **nine rows per cell line - the nine H3K79me2 samples**. No MEN1 or DOT1L sample appears in either table [E1][E2]. **Why this is the single fact that bounds the whole beat.** `share(s) = absolute(s)/G` is an exact identity. Without a spike-in, `G` - the genome-wide total the drug may have moved - is unknown, and every quantity computable from the peak tables is a within-sample contrast: - **MACS `fold_enrichment`** is ChIP pileup over the same input scaled to that ChIP's own depth. It reports signal-to-background **shape**. **If menin left every site by the same factor and the library were sequenced to the same depth, fold_enrichment would not move at all.** - **Peak recovery** is the same statement made binary. - A value above the DMSO value therefore means *'fell less than this sample's average'*, **not** *'gained'*. **What survives the bound, and it is not nothing.** A site going from a called peak at 3.7-6.2 fold-enrichment to no call at all, at 19 out of 19 pre-specified loci, while 1,101 peaks are still called elsewhere in the same library, is a **redistribution** statement and redistribution is exactly what a within-sample contrast can see. What cannot be said from this deposit is how much menin is on the genome in total under VTP-50469. **The remedy is one aliquot.** The depositors already run the dm6 protocol; adding fly chromatin to the MEN1 and DOT1L ChIPs would have converted this from a contrast to an absolute measurement at essentially no marginal cost. It is the cheapest unrealised improvement this beat found.
The one thing the shared input DOES buy
All ten factor ChIPs were called against a single input BAM (MOLM13_ChIP_input_DMSO_5d_191211_TGTATGCG_S13), verified by reading the control file line out of all ten MACS headers: one distinct value 2. That removes one whole class of confound - the arms are not each being divided by a different denominator - and it makes fold-enrichment comparable between samples in a way it usually is not. It still does not supply G, because the input is a DMSO input and is the same for every arm: it reports the genome, not the drug's effect on total occupancy.
| absence class | measured-not-reported |
|---|---|
| what is absent | dm6 spike-in counts and ratios for the ten MEN1 and DOT1L ChIP samples of GSE149183 |
| what is present | dm6 spike-in counts and ratios for the nine H3K79me2 samples per cell line (18 total) |
| what would close it | either a dm6 realignment of SRP258055's factor-ChIP runs if fly reads are present, or one repeat ChIP with fly chromatin added - the protocol the same lab already runs |
| consequence | every MEN1/DOT1L occupancy number from this deposit is a within-sample contrast; no absolute gain/loss statement is available |
| what a contrast CAN still show | redistribution - a site going from a called peak to no call while the library still calls 1,101 peaks elsewhere |
| shared input bam | MOLM13_ChIP_input_DMSO_5d_191211_TGTATGCG_S13_sorted_readgps_rmdup.bam |
| n distinct controls across ten chips | 1 |
| geo | GSE149183 |
| absence class note | FOLDED 2026-09-24: this entry originally carried the non-vocabulary token 'measured-not-deposited-for-this-assay', which is the truer description and is preserved here; 'measured-not-reported' is the nearest controlled token. |
depth range fragments
[ 11225637, 17750246 ]
- A pharmacodynamic endpoint built on a within-sample contrast can read as full target engagement while total occupancy has barely moved, or as no effect while half the protein has left the genome. Any occupancy-based PD marker for a menin inhibitor needs a spike-in or an orthogonal absolute readout before it goes anywhere near a trial.
- Do the SRP258055 factor-ChIP runs contain dm6-alignable reads? If the fly chromatin was in the tube and simply never aligned, the absolute scale is recoverable from raw data without any new experiment - and this deposit would become the only one in the corpus with an absolute menin-occupancy scale under two drug classes.
- How many occupancy claims in this corpus are within-sample contrasts presented without that qualifier? The identity share = absolute/G is exact, so the qualifier is never optional, and this beat needed it in every single entry.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.dataset
GSE149183's data-processing lines scope the Drosophila spike-in to the H3K79me2 samples, and the two deposited spike-in tables (MOLM13 and RS4_11, ...TSS-ChIPseq_signal_+_dm6_spike-in_...sitepro.txt.gz) each contain exactly nine spike-in rows, all H3K79me2, with none for MEN1 or DOT1L.
NCBI GEO GSE149183 - MEN1, DOT1L and H3K79me2 ChIP-seq in MOLM13 and RS4;11 under DMSO, VTP-50469, EPZ5676, compound 10 and compound 11; 31 samples, Homo sapiens; SubSeries of GSE149185, BioProject PRJNA627612, SRA SRP258055 · NCBI Gene Expression Omnibus 2020What this evidence cannot show
The tables were read directly from the deposited gzipped files, not from the series description. The MOLM13 table is comma-delimited and the RS4;11 table is tab-delimited - a formatting difference inside one deposit that will break a naive parser.
Evidence that argues against this (1)
- Absence of a spike-in COLUMN is not proof that no fly chromatin was added to the factor ChIPs; it proves no dm6 alignment was deposited for them. If the raw reads in SRP258055 contain fly sequence, the scale factor could be recomputed - that was not checked here and would require downloading the raw data this beat deliberately avoided. NCBI Gene Expression Omnibus 2020
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.dataset
All ten MEN1/DOT1L MACS2 headers name one and the same control BAM, so fold-enrichment across the ten samples shares a denominator; the headers carry per-sample treatment fragment counts (11.23-17.75 M) but no spike-in field.
analyses/menin-inhibitor-against-dot1l-inhibitors-in-one-deposit (PREREGISTRATION.md, run.py, results.json) - this repository, re-analysis of the ten deposited GSE149183 MACS2 peak tables · Oncofusion knowledge graph 2026What this evidence cannot show
A single shared input is a strength for cross-sample comparability and a weakness for detecting arm-specific chromatin accessibility changes, which a matched per-arm input would catch and this design cannot.
Evidence that argues against this (1)
- The shared input is a DMSO input taken at 5 d (191211) while the ChIPs are at 4 d from a different sequencing run (200109). Using a non-contemporaneous input for all arms is common practice and is still a source of systematic error that no arm-level comparison can remove. Oncofusion knowledge graph 2026
bounds every occupancy number in the headline result to a within-sample contrast - no absolute gain or loss is available
the ruler that entry found is deposited for the H3K79me2 samples only; the factor ChIPs in the same series are not in the table
adds the spike-in that would convert every contrast in this beat to an absolute scale