The only absolute-scale H3K79me2 deposit this project has contains no IgG or no-primary library, so the 1-31% residual measured after DOT1L inhibition cannot be separated from antibody background - the single cheapest missing control in the dataset
also called no IgG anywhere · the residual may be background · the detection floor is unmeasured
After nine days of the drug, genes still show one to thirty per cent of their original chromatin signal. Nobody knows whether that is real leftover mark or just the antibody sticking where it should not, because the experiment includes no control for that - and neither does any similar public experiment we could find.
The same thing, for a scientist
**The absence, bounded.** GSE149183 deposits 31 samples. Nineteen are ChIP (H3K79me2 x 18, plus MEN1 and DOT1L occupancy), three are chromatin inputs - GSM4491615 and GSM4491644 (MOLM13), GSM4491645 (RS4;11). **There is no IgG, no-primary-antibody or mock-IP library, and the inputs carry no dm6 spike-in table of their own.** **Why an input is not a substitute.** A chromatin input reports what fraction of the genome is accessible to sonication and sequencing. It does not report what the H3K79me2 antibody binds when H3K79me2 is absent. Those are different quantities and only the second bounds a residual signal. **What it costs, precisely.** Three results in this beat have a low-signal end that cannot be interpreted: | result | the part that is safe | the part that is not | |---|---|---| | retention gradient across deciles | the top: heavily marked genes fall below 1% retained | the bottom: 23-31% retained in the least-marked decile | | 'zero genes hold or gain' | holds at every threshold from 25 upwards | the 164 apparent holders at the loosest threshold | | the widening spread with time | the widening itself, seen in all four arms | whether the low tail is biology or floor | **And the same absence in the comparator deposits.** A survey of the H3K79me2 and spike-in entries already in this graph finds none carrying an antibody control either. This is a field-level habit, not a lapse by these depositors.
What would close it, in order of cost
- Computational, zero bench cost. siQ-ChIP-style analysis assesses histone antibody specificity directly from sequencing data, without a separate control library. If it can be run on the deposited alignments it would bound the floor from the existing deposit. Whether it can be run on
.tdftracks rather than raw reads is the open question; the raw data is in SRA under SRP258055. - Three libraries. A no-primary-antibody IP from the day-9 DMSO and both treated arms, with the same dm6 spike-in, read on the same TSS windows.
- The full replicated design, which includes this as one arm.
Option 1 costs a download and is not blocked by anything. It has not been attempted. 12
The reason this is worth an entry rather than a caveat
Because the residual is where the remaining biology is. A DOT1L inhibitor that removes 98-99% of H3K79me2 and produces modest clinical responses raises exactly one interesting question - what is the 1-2% that is left, and does it matter - and that question is unanswerable without knowing whether the 1-2% exists.
A caveat inside another entry would be read as a hedge. Stated as its own bounded absence, it is a specific, cheap, unclaimed task.
| script | analyses/absolute-h3k79me2-in-the-index-fusion/run.py |
|---|---|
| status | run |
| result summary | 31 samples; 19 ChIP, 3 chromatin inputs, 0 antibody controls. The 1-31% residual H3K79me2 after DOT1L inhibition therefore has no measured floor. |
| samples total | 31 |
| chip samples | 19 |
| input samples | 3 |
| antibody control samples | 0 |
| cheapest route to close | siQ-ChIP-style specificity analysis on the SRP258055 raw reads - a download, no bench work, unattempted |
| absence class | never-measured |
| absence class basis | GSE149183 deposits no IgG, no-primary-antibody or mock-IP library across 31 samples, so the floor was never measured rather than measured and withheld. |
inputs
[ "GSE149183 filelist.txt and SOFT family file" ]
input gsms
[ "GSM4491615", "GSM4491644", "GSM4491645" ]
- If the 1-2% residual at fusion target loci is antibody background, then DOT1L inhibition is complete at those loci and the failure of pinometostat cannot be blamed on incomplete target engagement - which removes the main argument for a better DOT1L inhibitor.
- If the residual is real, it is the most interesting 1% in this dataset and nobody has looked at it, because every analysis including this one normalises it away as noise.
- Can the siQ-ChIP specificity analysis be run on SRP258055 as deposited? That is the whole question and it costs a download.
- Is there ANY public spike-in H3K79me2 dataset with an antibody control? A systematic GEO query was not run for this beat and would be a clean small task.
- At day 9 the treated libraries are 98-99% depleted - close enough to a negative control that the residual could be modelled from them directly. Would that be circular, or is it the best available floor?
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.dataset
GSE149183's deposited sample list contains no IgG, no-primary-antibody or mock-IP library; its only non-ChIP samples are three chromatin inputs.
GSE149183 supplementary filelist.txt and SOFT family file (31 samples as served 2026-09-23) · Perner F, Gadrey JY, Xiong Y, et al. · NCBI Gene Expression Omnibus 2020What this evidence cannot show
This bounds what was DEPOSITED, not what was RUN. A control library may exist and not have been shared. The survey of comparator deposits was made against the H3K79me2 and spike-in entries already in this graph, not against a systematic search of GEO, so 'the same absence elsewhere' is a bounded statement about this corpus and not about the literature.
Evidence that argues against this (1)
- The raw sequencing data is public under SRP258055, so a background estimate could in principle be derived from the treated libraries themselves - at day 9 with 98-99% of the mark removed, the treated libraries are arguably the closest thing to a negative control the experiment contains, which weakens the claim that nothing bounds the floor. NCBI SRA 2020
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.preprint
Histone antibody specificity can be assessed directly in sequencing data without a dedicated control library, using siQ-ChIP.
Analysis of histone antibody specificity directly in sequencing data using siQ-ChIP. · Kupai A, Vaughan RM, Rothbart SB, Dickson BM · bioRxiv 2023What this evidence cannot show
Preprint, not peer-reviewed at the version cited, and read from its PubMed record and abstract rather than in full. Whether the method applies to this deposit's data form has not been checked.
Evidence that argues against this (1)
- A quantitative version of the same method was published in a peer-reviewed journal the same year, which strengthens the approach but does not establish that it recovers an antibody-background floor from an arbitrary third-party deposit. Scientific Reports 2023
the low-baseline end of the gradient is uninterpretable without a measured floor
the three no-primary libraries close this absence
the 1-31% residual after DOT1L inhibition has no measured floor
bounds the retention figures from below; the fold-down is a floor on the effect, not a ceiling