Oncofusion

start / graph / gene-chaf1a

Genemoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q37-histone-chaperones

CHAF1A (CAF-1 p150) - the subunit that reads PCNA and makes histone deposition replication-coupled, and the half of CAF-1 that leukaemia has not looked at

also called CHAF1A · CAF · CAF1P150 · CAF-1 p150 · chromatin assembly factor 1 subunit A · Q13111

CHAF1A is the part of the replication-coupled packaging machine that grabs the sliding clamp on the copying fork, which is what makes this route happen only while DNA is being copied. That coupling is the reason this graph's drug-timing argument has been framed in cell divisions.

The same thing, for a scientist

CHAF1A (Q13111, 956 aa) is the large p150 subunit of CAF-1, the complex that deposits H3.1/H4 behind the replication fork [E1]. **Its mechanistic distinction from CHAF1B is the one that matters for this beat: CHAF1A carries the DNA-binding winged-helix domain that functions with PCNA to stabilise CAF-1 at replication forks** [E2]. That interaction is the physical basis of 'replication-coupled', and therefore the physical basis of the assumption that H3K79 methylation is diluted per cell division. **CHAF1A has essentially no leukaemia literature of its own.** `(CHAF1A OR CHAF1B) AND (MLL OR KMT2A)` returns two records and both are CHAF1B papers [E3]; the leukaemia work is on p60. In DepMap CHAF1A is deeply pan-essential (-1.206) with a difference of **+0.073** - marginally LESS required in KMT2A-rearranged lines [E4]. This entry exists for two reasons: the p150/PCNA coupling is load-bearing for the clearance-clock argument, and **the p150/p60 asymmetry in the literature is itself a finding** - the CAF-1 subunit with the leukaemia biology is not the one with the fork-coupling mechanism.

Go deeper
The two subunits are not interchangeable, and the alias system says they are

CHAF1B carries the curated alias symbol CAF1A while the actual CHAF1A gene is a different record of nearly twice the length (956 aa against 559 aa). That is rule 63's trap, in this graph's own subject matter: a lookup that matches alias fields can resolve the string CAF1A to the p60 subunit.

Both accessions are asserted in these entries so a length check can catch a swap:

subunitsymbolUniProtlengthcurated aliases
p150CHAF1AQ13111956CAF, CAF1P150
p60CHAF1BQ13112559CAF1A, CAF1P60, MPHOSPH7, MPP7
p48RBBP4Q09028425RBAP48

And because CAF-1 is usually discussed by complex name rather than by subunit symbol, the complex-name rule applies too: a subunit-only enumeration will miss papers that frame their result at the level of 'CAF-1'.

Facts and figures
uniprotQ13111
gene symbolCHAF1A
length AA956
complexCAF-1 (CHAF1A p150, CHAF1B p60, RBBP4 p48)
deposition modereplication-coupled
mechanismDNA-binding winged-helix domain functions with PCNA to stabilise CAF-1 at replication forks
kmt2a literature records0
alias hazardCAF1A is a curated alias symbol of the OTHER subunit CHAF1B (Q13112, 559 aa)
depmap
{
  "index_mean": -1.2194,
  "other_leukaemia_mean": -1.2929,
  "difference": 0.0735,
  "pan_essentiality": -1.2063,
  "n_index": 11,
  "interpretation": "floor-compressed; not interpretable as selectivity"
}
depmap kmt2a selectivity
{
  "chronos_mean_kmt2a_rearranged": -1.2194,
  "chronos_mean_other_leukaemia": -1.2929,
  "difference": 0.0735,
  "pan_essentiality_all_lines": -1.2063,
  "selectivity_rank": 16119,
  "rank_denominator": 17728,
  "provenance": "Joined 2026-09-21 from analyses/depmap-kmt2a-selective-scan-full (generated 2026-09-19T10:12:06Z), the project's own genome-wide scan over 11 KMT2A-rearranged and 43 other leukaemia DepMap lines. NOT re-fetched and NOT independent of that scan: every caveat it carries applies here, including that the KMT2A-rearranged arm is a small number of lines drawn from nine individuals, that the rank denominator is 17,728 genes, and that a difference is not a dependency. Added because only 26 of 308 gene entries carried this number, which made any KMT2A-selectivity question answerable for 8% of the corpus.",
  "how_to_read": "A negative difference means MORE required in KMT2A-rearranged lines than in other leukaemia lines. A rank near 17,728 is ANTI-selective. Neither is evidence of a fusion-driven dependency on its own: check pan_essentiality_all_lines before calling anything selective, and note that a paralogue-redundant gene can score near zero for reasons that have nothing to do with this disease - see ana-the-paralogue-redundancy-excuse-is-not-tested-and-my-proxy-decided-the-answer."
}
How this could bite us back
  • CHAF1A is required to package every newly replicated genome; DepMap scores it at -1.21 across 1,208 lines and it has no selectivity for this disease, so a complete block has no window.
  • CAF-1 loss channels replication-fork problems into recombination-dependent outcomes, so partial inhibition carries a genome-instability risk in a paediatric patient that no leukaemia trial endpoint measures.
  • Blocking replication-coupled deposition slows the per-division dilution of parental histones, which by this graph's clearance-clock argument makes H3K79 methylation persist longer, not shorter.
Still open
  1. Is the leukaemia phenotype of CAF-1 a p60 phenotype or a complex phenotype? Every published result is on CHAF1B and nobody has tested whether CHAF1A reduction does the same thing.
  2. Re-run the CAF-1 absence against the complex name 'chromatin assembly factor 1' as a quoted phrase rather than against subunit symbols; this project has twice found that a complex-level framing is unreachable from subunit symbols.
  3. How much of H3K79me2 loss under DOT1L inhibition is CAF-1-dependent dilution versus replication-independent turnover? A CHAF1A-degron in a KMT2A-rearranged line would separate them directly and has never been run.
Every source behind this entry (5)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    The purified H3.1 complex contains CAF-1 and mediates DNA-synthesis-dependent nucleosome assembly, whereas the H3.3 complex contains HIRA and mediates DNA-synthesis-independent assembly; the two chaperones define the two deposition pathways.

    PMID 14718166DOI
    What this evidence cannot show

    Abstract read from efetch 2026-09-20; full text not read. Tagged-histone purification from HeLa; the complexes recovered are those stable enough to co-purify with an over-expressed tagged histone. The abstract speaks of CAF-1 as a complex and does not resolve the contributions of p150, p60 and p48 individually.

    Evidence that argues against this (1)
    • Reconstitution of human CAF-1-dependent nucleosome assembly in a defined system shows the activity can be studied outside a replicating nucleus, so 'replication-dependent' describes recruitment and efficiency rather than an absolute requirement of the enzymology. Kadyrova LY et al. 2013
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    A DNA-binding winged-helix domain in CAF-1 functions with PCNA to stabilise CAF-1 at replication forks.

    PMID 26908650DOI
    What this evidence cannot show

    Record fetched from efetch 2026-09-20 for identifiers and title; abstract read, full text not read. **The domain is attributed to 'CAF-1' in the title and this entry attributes it to the p150 subunit, which is the standard assignment but is an attribution this entry did not verify from the paper's own text.** No residue span is stored for that reason - this project's rule is that a domain boundary must be derived by alignment to a freshly fetched UniProt sequence, and it was not.

    Evidence that argues against this (1)
    • Yeast work on the Cac1 winged-helix and PIP domains separates their roles in replisome progression from their roles in genomic stability, so a single 'stabilises CAF-1 at forks' statement collapses at least two distinguishable functions. Tsirkas I et al. 2021
  3. E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    As of 2026-09-20, PubMed returns 2 records for '(CHAF1A OR CHAF1B) AND (MLL OR KMT2A)' and on title check both are CHAF1B papers - the 2018 Cancer Cell molecular-switch study and the 2023 Blood Advances TRIM13 study. There is no CHAF1A-specific KMT2A or MLL record.

    What this evidence cannot show

    Query translation captured; no [UID] term. Both titles fetched in full. **This absence is bounded by subunit symbols and this project has documented that a complex-level framing is unreachable from subunit symbols** - the MED23 and KANSL cases. A paper discussing 'CAF-1' in a KMT2A model without naming CHAF1A would not be caught, and the complex-name query was not run in this form.

    Evidence that argues against this (1)
    • Searching a complex by its subunit symbols under-counts by construction: the paper establishing that five NUP98 fusions bind the NSL complex appears in no query built from KANSL1, KANSL2 or KANSL3. The same failure is likely here and this absence should be re-run against the phrase 'chromatin assembly factor 1'. Dean ST et al. 2023
  4. E4moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset

    In DepMap CRISPR gene effect, CHAF1A's mean effect is -1.219 in 11 KMT2A-rearranged leukaemia lines against -1.293 in 43 other leukaemia lines, a difference of +0.073 (slightly LESS required in the index arm), at a pan-essentiality of -1.206 across 1,208 lines; the p48 subunit RBBP4 sits at -1.347 pan-essentiality with a difference of -0.024.

    What this evidence cannot show

    Run 2026-09-20. At a pan-essentiality of -1.21 this is floor-compressed and the difference is not interpretable as selectivity in either direction. Reported so that the absence of a signal is on the record rather than being rediscovered.

    Evidence that argues against this (1)
    • All three CAF-1 subunits are deeply pan-essential in CRISPR, yet halving CHAF1B in a mouse is tolerated and anti-leukaemic. A knockout screen cannot resolve a partial-occupancy window, so a CAF-1 subunit's pan-essentiality is not a verdict on whether CAF-1 can be drugged. Volk A et al. 2018
  5. E5established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    CHAF1A is UniProt Q13111 (CAF1A_HUMAN), chromatin assembly factor 1 subunit A / CAF-1 subunit A, 956 amino acids, with alias symbols CAF and CAF1P150; the p60 subunit CHAF1B is a distinct record Q13112 of 559 amino acids that carries CAF1A among its own alias symbols.

    UniProtKB Q13111 (CAF1A_HUMAN), Chromatin assembly factor 1 subunit A, 956 aa · UniProt Knowledgebase, reviewed (Swiss-Prot) 2026
    What this evidence cannot show

    Both records fetched 2026-09-20 with `gene_primary` requested and read back per rule 62; each query returned exactly one reviewed human record whose primary symbol matched. The alias collision was found in the returned synonym fields, not assumed.

    Evidence that argues against this (1)
    • This is the CSNK2A1/CSNK2A2 pattern in a different complex: the wrong answer is a same-complex subunit of a plausible size and nothing about the returned record reads as an error. Only the asserted accession and length catch it. UniProt Knowledgebase, reviewed (Swiss-Prot) 2026
Connected to