Oncofusion

start / graph / ana-q199-the-protein-atlas-calls-aff1-mitochondrial-and-nucleoplasmic-on-one-approved-tier-antibody-and-the-table-does-not-say-which-cell-line

Analysismoderate confidence 3 of 5: One strong study, or several smaller consistent ones.

The Human Protein Atlas lists AFF1 as mitochondrial and nucleoplasmic at only 'Approved' reliability, with mitochondria as the single-cell-variable compartment, against nucleoplasm for KMT2A, menin and most of the pathway - a localisation lead that needs an orthogonal check

also called AFF1 mitochondria · AF4 localisation Human Protein Atlas

A public protein atlas places the leukaemia fusion's partner protein partly in the mitochondria, unlike its nuclear partners. The evidence is a single lower-confidence antibody, so it may well be an artefact.

The same thing, for a scientist

HPA record for AFF1: reliability Approved; main location Mitochondria;Nucleoplasm; single-cell variation in mitochondria. For scale, 7.2% of all 13597 proteins have a mitochondrial main location, and 6993 (51%) carry the same 'Approved' tier, the second-lowest of four (Enhanced > Supported > Approved > Uncertain). The other pathway proteins are nucleoplasmic at Supported or Enhanced tiers (KMT2A, MEN1, DOT1L, MLLT1/3/10, PSIP1).

Go deeper
Why this is a lead and not a finding

Mitochondrial signal is a recognised immunofluorescence artefact, and 'Approved' is the tier with the least validation. The downloaded table has no cell-line column, so the cell type behind this call cannot be read from it. This corpus has circular-RNA work on mitochondrial restraint in KMT2A::AFF1 cells, which concerns a different molecule and must not be conflated with AFF1 protein.

Facts and figures
scriptanalyses/q199-germline-genetics-of-the-fusion-pathway/run.py
mito main share0.072
reliability counts
Uncertain792.0Uncertain: 792Enhanced1,230Enhanced: 1230Supported4,582Supported: 4582Approved6,993Approved: 6993
aff1 hpa
{
  "gene": "AFF1",
  "in_hpa": true,
  "reliability": "Approved",
  "main": "Mitochondria;Nucleoplasm",
  "additional": null,
  "ccd": null
}
reliability counts — the numbers behind the chart
{
  "Approved": 6993,
  "Supported": 4582,
  "Enhanced": 1230,
  "Uncertain": 792
}
How this could bite us back
  • If AFF1 had a real mitochondrial role, the fusion's 3-prime half might engage mitochondria too, with metabolic consequences; this is untested and the evidence is thin.
Still open
  1. Does AFF1 (or the AF4 N-terminus retained in the fusion, which is the part the fusion actually carries) localise to mitochondria in a KMT2A-rearranged line? The fusion retains AFF1 residues 347 onward, so an antibody to the lost N-terminal region would not see it.
Every source behind this entry (2)
  1. E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database

    The Human Protein Atlas lists AFF1 at 'Approved' reliability with main location 'Mitochondria;Nucleoplasm' and single-cell variation in the mitochondrial compartment.

    What this evidence cannot show

    Read from the downloaded subcellular_location.tsv (v25, Last-Modified 2025-11-06). The reliability tiers in this table are Enhanced, Supported, Approved and Uncertain.

    Evidence that argues against this (1)
    • Mitochondrial staining is a common artefact of immunofluorescence, and the Approved tier means limited antibody validation. The table carries no cell-line information, so the call cannot be tied to a leukaemia cell, and a nuclear fraction of AFF1 is also listed. A validated second antibody or a tagged AFF1 in a KMT2A-rearranged line showing no mitochondrial signal would refute it. Human Protein Atlas 2025
  2. E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database

    Across 13597 proteins in the same table, 7.2% have a mitochondrial main location, so AFF1's call is unusual for a transcription-elongation scaffold but not for the table as a whole.

    What this evidence cannot show

    Baseline computed from the Main location column; a mitochondrial annotation anywhere (additional location included) was not counted.

    Evidence that argues against this (1)
    • A base rate of about 7% for a mitochondrial main location does not make AFF1's call implausible; it only calibrates how surprising a mitochondrial call is for an arbitrary protein. Human Protein Atlas 2025
Connected to