Oncofusion

start / graph / absence-q120-not-one-of-gse260742s-158-sample-records-names-an-antibody

Analysisstrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q120-h3k79me2-second-line-and-the-mll1-epitope

A ChIP-seq series of 158 samples that names no antibody anywhere in its metadata: the string 'antibod' occurs ZERO times in GSE260742's 419,824-character SOFT dump, and the reagents are in a supplementary spreadsheet the full text never quotes

also called GSE260742 names no antibody · the empty antibody field

A public dataset of 158 chromatin experiments does not record which antibody was used for any of them. The information exists, but only in a spreadsheet attached to the paper, which no database search can reach.

The same thing, for a scientist

**Measured, not asserted** [E1]. `acc.cgi?acc=GSE260742&targ=gsm&form=text&view=brief` returns 158 `^SAMPLE` records in 419,824 characters. A case-insensitive count of `antibod` over that text returns **0**. Individual records carry `!Sample_title` (e.g. *'MLLAF9 Control - DMSO MLL1'*), `!Sample_characteristics_ch1` (tissue, cell line, cell type, treatment duration) and five `!Sample_data_processing` lines, and no reagent field of any kind. **The contrast inside this project's own evidence base is instructive.** GSE127507, deposited in 2019 by a different group, carries `antibody: anti-MLL1n (Bethyl #A300-086A lots: 5 or 6)` as an explicit per-sample characteristic on every one of its 26 samples, lot numbers included [E2]. So this is not a limitation of GEO's model - it is a per-submission choice, and the 2025 deposition is the poorer one. **Where the answer actually is.** The article's Methods say only *'All antibodies used in this study are commercially available polyclonal antibodies that have been validated per manufacturer's websites'*, with no catalogue numbers. The reagents are in `41556_2025_1859_MOESM6_ESM.xlsx`, a 24,492-byte eight-sheet key-resources workbook inside the 42.9 MB Europe PMC supplementary archive, first sheet `Antibody`, 27 rows. **Why it is load-bearing here rather than merely untidy.** The single fact that decides how this beat's mouse arm may be read - that the MLL1 ChIP uses the same Bethyl A300-086A as the human MLL1n track - is in that sheet and nowhere else that a database query reaches. A reagent audit built on GEO metadata would have scored this series as having no antibody information and stopped.

Go deeper
The bound, and what it does not cover

What is bounded: the !Sample_* records of GSE260742, all 158, fetched in one request on 2026-09-23, searched case-insensitively for antibod. Zero hits. Also searched: the series record (targ=self) and the article's full text via Europe PMC fullTextXML (268,228 bytes, <body> present) - the full text mentions antibodies in prose and names no ChIP catalogue number.

What is NOT bounded: the SRA sample attributes (SRX23818778 and siblings) were not fetched, and BioSample records (SAMN40239771 and siblings) were not fetched. Either could in principle carry an antibody attribute. Bound this as 'of the GEO series and sample records and of the article full text', which is what was searched.

Facts and figures
statusrun
absence claimNo GEO sample or series record in GSE260742 names any antibody, and the article's Methods name no ChIP catalogue numbers.
date measured2026-09-23
n records searched158
characters searched419824
hits0
what closes itIt is already closed, by the paper's supplementary key-resources workbook. The absence is in the DATABASE, not in the literature.
absence classmeasured-not-curated
absence class notePublished but absent from the database this project joins against. GEO supports a per-sample antibody characteristic - the comparison series GSE127507 fills it on 26 of 26 samples with lot numbers - and this submission left it empty on 158 of 158 while the reagents sat in a supplementary spreadsheet. Structurally invisible to any metadata-driven audit; closes by reading the supplement.
absence class assigned2026-09-23 (Q120)
strings tried
[
  "antibod",
  "Bethyl",
  "A300",
  "Abcam",
  "Cell Signaling",
  "catalogue"
]
routes tried
[
  "acc.cgi targ=gsm",
  "acc.cgi targ=self",
  "Europe PMC fullTextXML",
  "Europe PMC supplementaryFiles (SUCCEEDED)"
]
routes not tried
[
  "SRA sample attributes (SRX23818778)",
  "BioSample (SAMN40239771)"
]
How this could bite us back
  • An automated pipeline that scores public ChIP data by target label rather than by reagent will silently pool an N-terminal and a C-terminal antibody, or a fusion-blind and a fusion-specific one, under the same 'MLL1' heading. This series makes that unavoidable from metadata alone.
Still open
  1. How common is this? A count of empty antibody fields across every ChIP series this project cites would say whether GSE260742 is unusual or whether the 2019 series is the unusual one.
  2. Do SRA sample attributes carry antibody information for this series? One elink and one fetch would tighten the bound and were not run.
Every source behind this entry (2)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database

    The string 'antibod' occurs 0 times, case-insensitively, in the 419,824-character SOFT dump of all 158 GSE260742 sample records fetched 2026-09-23; the article (PMID 41634341) names no ChIP antibody catalogue number in its Methods; the reagents are in supplementary file 41556_2025_1859_MOESM6_ESM.xlsx sheet 'Antibody', 27 rows.

    What this evidence cannot show

    Bounded to the GEO series and sample records and to the article full text. SRA sample attributes and BioSample records were not fetched and could carry the field. The supplementary figures PDF (MOESM1, 7 pages) is image-only with zero extractable characters and zero form fields, so anything stated only there is outside this bound.

    Evidence that argues against this (1)
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database

    GSE127507, by contrast, carries an explicit per-sample antibody characteristic with catalogue and lot numbers on all 26 samples, e.g. 'antibody: anti-MLL1n (Bethyl #A300-086A lots: 5 or 6)' and 'antibody: anti-H3K79me2 (Cell signaling #5427 lot:4)'.

    What this evidence cannot show

    Read directly from the records; no inference.

    Evidence that argues against this (1)
Connected to