GSE127507 deposits a matched input for BOTH conditions in MOLM13 only - RS4;11 and ML-2 have a DMSO input and no VTP-50469 input - so no input-corrected cross-condition occupancy statement is possible in the human index-fusion line from this series, and none exists anywhere
also called the RS4;11 input asymmetry · no VTP input for the index fusion
To be sure a change in a treated sample is real you need the matching background control for the treated cells, and this dataset only has it for one of its three cell lines - not the one that matters most here. Every number for the index-fusion line therefore rests on a statistic that does not need it, which is a real constraint and not a formality.
The same thing, for a scientist
**Counted directly from the 26 sample records** [E1]: | line | fusion | DMSO input | VTP-50469 input | |---|---|---|---| | RS4;11 | **KMT2A::AFF1 - the index fusion** | GSM3635699 | **NONE** | | MOLM13 | KMT2A::MLLT3 | GSM3635689 | GSM3635690 | | ML-2 | KMT2A::AFDN | GSM3635680 | **NONE** | **Two consequences, and the second is the one that constrains this beat.** 1. **The depositors' own processed table inherits it.** The ROSE TSS signal table is *"input-subtracted ... input samples matched by cell line"*, so the **RS4;11 and ML-2 VTP columns are input-subtracted with a DMSO input.** That cannot correct for any treatment-induced background change, and it cannot be repaired from the deposit. 2. **No differential-input-corrected cross-condition statement is available for the index-fusion line at all**, from this series or from any other known to this graph. This beat's index-fusion numbers are therefore within-sample shares and within-sample ranks exclusively - statistics that do not require an input in either condition - and the one input-corrected figure in the beat (MLL1-N held-set share ratio 0.9438 correcting to 0.9891) is from **MOLM13**, transferred to RS4;11 only as a stated bound and never as a measurement. **Measured size of the correction where it can be measured.** In MOLM13 the input's own within-sample share at the MLL1-N held set moves by **4.6%** (ratio 0.9542). So an input correction of that magnitude is **immaterial to a 5-fold effect** (the RS4;11 menin collapse, 0.1999) and **material to a 10% one** (the RS4;11 MLL1-N figure, 0.9069, which would correct to roughly 0.95 if RS4;11's input behaved like MOLM13's). **The retention claims in this beat are the ones this absence bears on, not the eviction claims.** **This is a `measured-not-reported` absence and it is closeable by wet work only.** The raw reads exist (SRP187096 / PRJNA524887) but a sample that was never sequenced cannot be recovered from an archive. **The cheap partial substitute** is to use the MOLM13 VTP input as a background model for the RS4;11 VTP ChIP, which is wrong in a known direction (different karyotype, different copy number) and would at least bound the effect. It was not done here.
| absence class | measured-not-reported |
|---|---|
| absence class note | The gap this token describes is the RS4;11 and ML-2 VTP-50469 INPUT samples specifically. The assay generates an input for every ChIP experiment as a matter of routine, so one was very likely prepared and it was not deposited; SRP187096 was not enumerated to check. This entry carries only that one gap. It does NOT describe the tagged-fusion experiment (never-measured in human) or the H3K79me2 analysis (measured-not-analysed). |
| what would close it | Either an undeposited RS4;11 VTP input recovered from SRP187096 or from the depositors, or a new RS4;11 VTP-50469 input library - one sequencing lane. |
| cheap partial substitute | use the MOLM13 VTP input as a background model for the RS4;11 VTP ChIP; wrong in a known direction (different karyotype and copy number) but bounds the effect. Not done here. |
| series | GSE127507 |
| sra study | SRP187096 |
| bioproject | PRJNA524887 |
| measured correction size where available | 4.6% on the input's own within-sample share at the MLL1-N held set in MOLM13 |
| script | analyses/q118-menin-at-the-human-index-fusion/PREREGISTRATION.md section 2 (input asymmetry) |
| status | run |
| result summary | Of 26 samples, four are inputs; MOLM13 alone has one in both conditions. The RS4;11 and ML-2 VTP columns of the depositors own input-subtracted table therefore use a DMSO input. |
inputs deposited
[ "GSM3635680 ML-2 DMSO", "GSM3635689 MOLM13 DMSO", "GSM3635690 MOLM13 VTP", "GSM3635699 RS4;11 DMSO" ]
- Any published claim from this dataset about RS4;11 occupancy under VTP-50469 that used an input subtraction used a DMSO input for the treated sample. That includes the depositors' own processed table, which is the file most re-users will open.
- Does an undeposited RS4;11 VTP-50469 input exist under SRP187096 or PRJNA524887? The SRA study was not enumerated in this beat and this project has documented three separate routes that falsely report a SuperSeries as having no raw data. One SOFT family-file read would settle it.
- Would the MOLM13 VTP input work as a background model for RS4;11? It is wrong in a known direction and would bound the correction. Both files are already binned and on disk; it is one line of code and it was not run.
- How many other menin-inhibitor occupancy datasets have this asymmetry? This graph holds absence-no-menin-inhibitor-occupancy-dataset-in-an-infant-derived-kmt2a-aff1-line, so the question of what inputs exist across the whole field has not been surveyed.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.dataset
Of the 26 GSE127507 samples, four carry antibody: none and are inputs - GSM3635680 (ML-2 DMSO), GSM3635689 (MOLM13 DMSO), GSM3635690 (MOLM13 VTP-50469) and GSM3635699 (RS4;11 DMSO) - so MOLM13 is the only line with an input in both conditions, and the depositors' own input-subtracted TSS table therefore subtracts a DMSO input from the RS4;11 and ML-2 VTP columns.
NCBI GEO GSE127507 - MENIN, MLL1n, DOT1L and H3K79me2 ChIP-seq in RS4;11, MOLM13 and ML-2 with and without VTP-50469; 26 samples, Homo sapiens, hg19; SubSeries of GSE127508 · NCBI Gene Expression Omnibus 2019What this evidence cannot show
This is a statement about the deposited samples. A VTP input may have been sequenced for RS4;11 and not deposited, in which case the class would be measured-not-reported in a stronger sense and a request to the depositors could recover it. The SRA study SRP187096 was NOT enumerated in this beat, so the possibility that an undeposited input exists under that BioProject has not been checked - and this project has documented that a SuperSeries reports no raw data three different ways and all three lie.
Evidence that argues against this (1)
- The MOLM13 measurement shows the correction is small (4.6% on the input's own share at the held set), so the practical cost of this absence to this beat's headline numbers is low. It is reported as a hard constraint anyway because the retention figures - the 0.91 that the whole attribution argument turns on - are in the range where a 4.6% correction matters proportionally. Iqbal Lab oncofusion knowledge graph 2026
no differential-input correction is possible for RS4;11; the correction size is transferred from MOLM13 as a bound
MOLM13 is the exception to the input asymmetry and supplies the only differential-input-corrected figure in the series (0.9891)
adds a second structural gap in the human menin-inhibitor occupancy record: not only no infant-derived KMT2A::AFF1 dataset, but no treated-condition input for the adult-derived one that exists
RS4;11's DMSO input is GSM3635699 and there is no VTP-50469 input
the design includes an input in both conditions, which is the gap this absence records
two deposits each nearly answer the index-fusion question and each lacks a different single sample type