BGPT: Paper Review: Phylogenetic distribution of DNA topoisomerase VI and its distinction from SPO11

Fuel Your Discoveries

Quick Explanation Copied

Paper-focused core verdict

Allen & Maxwell provide a large, phylogeny-driven framework to distinguish topo VI-A vs Spo11 and to separate topo VI-B vs topo VIB-like using a combination of conserved-residue logic, maximum-likelihood clustering, and AlphaFold-based structural discriminants. They argue that Apicomplexa (including Plasmodium) lack a bona fide topo VI complex because the Apicomplexa “putative TOP6B” appears structurally/phylogenetically closer to topo VIB-like rather than TOP6B, resolving prior inconsistencies about Plasmodium topo VI presence.

Main evidence is computed/sequence-based (no direct enzymology in Plasmodium in this paper).

Source:

Long Explanation

Phylogenetically mapping topo VI vs Spo11 (and testing the Plasmodium claim)

NARGAB (published 2024-08-06). DOI: 10.1093/nargab/lqae085

1) Visuals first: what the paper actually built

2) What the paper claims (with skeptical framing)

2.1 Distinguishing topo VI-A vs Spo11 (sequence-level, phylogeny-level)

Methodic core: they align and build unrooted maximum-likelihood trees for a curated set spanning Archaea/Bacteria/Eukaryota; they report that topo VI-A/Spo11 splits into three major clades (topo VI-A/SPO11-3, SPO11-1, SPO11-2).
Key biological interpretation: they propose a duplication history separating SPO11-3 from SPO11-1 first, followed by separation of SPO11-1 and SPO11-2 later in eukaryotic evolution.

Skeptical check: clade clustering is powerful for orthology assignment, but it can be confounded by long-branch effects, alignment/model mismatch, and sampling gaps—especially because deep evolutionary relationships are sensitive to phylogenetic artifacts. The paper uses trimmed alignments and large bootstrap counts, but the interpretability of ancient duplication/HGT timings still depends on model adequacy and taxon coverage.

2.2 Distinguishing topo VI-B vs topo VIB-like (sequence alone is not enough)

Sequence limitation acknowledged: they argue topo VI-B and topo VIB-like are not reliably separable by sequence alignment/phylogeny alone due to limited homology.
Structural discriminant: they use AlphaFold2/ColabFold structural modeling and compare predicted presence/absence of a key domain feature (H2TH) and transducer/GHKL architecture, concluding topo VIB-like lacks H2TH and has a different transducer-like geometry.

Skeptical check: AlphaFold-based domain presence/absence is suggestive, but it is still predictive: a domain could be mispredicted under constraints, or a divergent homolog could fold differently than expected. Without experimental structures or biochemical validation across the major “ambiguous” clades, this remains an informed inference rather than direct proof.

2.3 The headline conclusion: Apicomplexa/Plasmodium likely lacks topo VI

Phylogenetic claim: Apicomplexa “pTOP6B” sequences form a separate clade that the authors interpret as distantly related to eukaryotic TOP6B.
Structural claim: predicted pTOP6B lacks a full H2TH domain and is modeled as belonging to topo VIB-like rather than true TOP6B.
Accessory protein logic: in their SAR mapping, presence of topo VI is tied to the accessory proteins RHL1 and BIN4 in topo VI-possessing eukaryotes, and they report these are absent in Apicomplexa.

Where this is strongest: combining (i) phylogenetic distinctness, (ii) structural feature prediction, and (iii) accessory-protein co-occurrence is a multi-evidence argument that is more robust than any single axis.

Most important remaining uncertainty: functional status of the predicted Apicomplexa protein complex(s) is not established experimentally here; the main falsification would be finding bona fide topo VI enzymology or a true TOP6B/TOP6A-like heterotetramer in Apicomplexa.

3) Method critique (what could systematically mislead)

Homology-based ID risk: the central inference is assignment by phylogenetic/structural discriminants; misannotation or deep divergence could yield wrong identity calls. The paper explicitly addresses this by building curated datasets and using domain logic (e.g., conserved catalytic tyrosine filtering and separate treatment of topo VI-B vs VIB-like), but the risk cannot be eliminated entirely without experimental ground truth.
Phylogeny artifacts: unrooted ML topology can be sensitive to alignment trimming choices and model mismatch; while they use extensive bootstrapping, deep events like “HGT number of transfers” are still probabilistic.
Structural prediction uncertainty: AlphaFold-style outputs are probabilistic; they are best treated as hypotheses about domain architecture rather than definitive structural assignment. The paper uses known crystal structures for context and compares predicted architectural hallmarks (e.g., H2TH presence/absence).
Genome incompleteness & contamination: distribution statements in protists can be distorted by assembly quality and bacterial contamination; the paper itself discusses an exception lineage (Cladocopium goreaui) and contamination possibility in its discussion.

4) What would most credibly falsify the paper’s Plasmodium topo VI conclusion?

Direct biochemical/enzymatic proof of a canonical topo VI complex (including the right subunit pairings consistent with topo VI-A clade assignment and topo VI-B/TOP6B architecture) in Apicomplexa, rather than only predicted/sequence-based identifications.
Re-derivation under stronger controls: alternative orthology pipelines and structural class predictors that independently assign pTOP6B as true TOP6B would weaken the paper’s structural discriminant.

BGPT next-step buttons

Feedback:

Updated: March 30, 2026