AIGR Deep Research — csr-1 (Q21992): predicted DNA-binding TF activity & metabolic-process regulation
Gene: csr-1 (C. elegans, NCBITaxon:6239) | Seed accession: Q21992 Hypothesis (BioReason-Pro SFT): DNA-binding transcription factor activity (GO:0003700) and regulation of metabolic process (GO:0019222; positive regulation GO:0009893).
Executive Judgment
Verdict: REFUTED (with a critical accession-mismatch caveat).
The prediction of DNA-binding transcription factor activity (GO:0003700) is refuted under both plausible identities of the supplied identifier:
- The gene named
csr-1(WormBase locus F20D12.1) encodes the CSR-1 Argonaute — a small-RNA (22G-RNA)-guided effector with PAZ (IPR003100) + PIWI (IPR003165 / Pfam PF02171) + RNaseH-like domains. It is a post-transcriptional/chromatin-associated RNA slicer, not a sequence-specific DNA-binding transcription factor. Current UniProt accessions: H2KZD5 (CSR-1a, 1030 aa) and Q27GU1 (CSR-1b, 867 aa). - The supplied accession Q21992 is DELETED in UniProt (Inactive/DELETED; UniParc UPI000007665A). Its sequence maps unambiguously to a different gene — R144.7 =
larp-1(La-related protein 1), an RNA-binding translational regulator (current reviewed entry D5MCN2, 1150 aa; RefSeq NP_001040868.3 "La-related protein 1"). LARP-1 binds RNA (poly-U/poly-G, RNA cap) and positively regulates translation — it is also not a DNA-binding transcription factor.
Neither identity supports GO:0003700. The metabolic-process regulation terms (GO:0019222 / GO:0009893) are so broad that they are technically not false for either gene (both regulate a macromolecule-metabolic process — gene silencing or translation), but they are uninformatively broad and non-core, and they do not rescue the mechanistic TF claim, which is the substantive part of the prediction.
Most important caveat: There is a genuine identifier problem. The seed pairs symbol csr-1 with accession Q21992, but Q21992 ≠ CSR-1. A curator must first resolve which protein is intended. Under either resolution the TF prediction fails.
Sequence-Level Provenance (computed this run)
Direct pairwise 6-mer containment on retrieved sequences resolves the identity quantitatively:
| Comparison | Lengths | 6-mer containment | Interpretation |
|---|---|---|---|
| Q21992 (deleted seq) vs LARP-1 (D5MCN2) | 1065 vs 1150 | 0.971 | Same protein (identical N-terminus MAEKQPMLSFAKVVSGQAED...); Q21992 is an older LARP-1 isoform |
| Q21992 vs CSR-1a Argonaute (H2KZD5) | 1065 vs 1030 | 0.000 | Unrelated — no shared 6-mers |
| LARP-1 (D5MCN2) vs CSR-1a (H2KZD5) | 1150 vs 1030 | 0.000 | Unrelated proteins |
This confirms the seed's csr-1 → Q21992 pairing is an accession error: Q21992 is larp-1, sharing zero detectable sequence with the CSR-1 Argonaute.
Evidence Matrix
| Citation | Evidence type | Supports/Refutes | Claim tested | Key finding | Context | Confidence / limitations |
|---|---|---|---|---|---|---|
| UniProt H2KZD5, Q27GU1 (database) | Structural/domain | Refutes GO:0003700 | csr-1 is a DNA-binding TF | CSR-1a/b carry PAZ + PIWI + RNaseH-like; no DNA-binding/TF domain | C. elegans, protein family | High; database-level but based on curated domain models |
| 19804758 (Claycomb 2009) | Mutant phenotype / localization | Refutes | csr-1 identity/function | CSR-1 Argonaute + EGO-1/DRH-3/EKL-1 localize to chromosomes and are required for chromosome segregation via 22G-RNAs; CSR-1 does not down-regulate target mRNA/protein | C. elegans germline/embryo | High; primary paper |
| 24178449 (Wedeles 2013, review) | Review synthesis | Refutes | csr-1 identity | CSR-1 binds 22G-RNAs complementary to ~25% of genes; only essential Argonaute of 24 | C. elegans | High for identity; review-level |
| 22863779 (Avgousti 2012) | Direct assay | Qualifies (BP) | csr-1 regulates gene expression | CSR-1 directly binds histone mRNA; RNAi pathway positively regulates histone expression | C. elegans germline | High; shows post-transcriptional (not TF) regulation |
| 25510497 (Tu 2015) | Comparative genomics | Refutes | csr-1 identity | Conserved nuclear Argonaute role in chromatin/segregation/germline expression across Caenorhabditis | C. elegans / C. briggsae | High |
| UniProt Q21992 record (database) | Database | Competing/refutes | Q21992 = csr-1 | Q21992 is DELETED; UniParc maps to R144.7 (larp-1), RefSeq NP_001040868 = "La-related protein 1" | Identifier check | High; direct record retrieval |
| UniProt D5MCN2 (database) | Structural/domain | Refutes | Q21992-seq is a DNA-binding TF | LARP-1: La domain (IPR045180), DM15, La-HTH; only a winged-helix-like structural fold (IPR036388); function = RNA binding + positive regulation of translation | C. elegans | High; La proteins bind RNA, not DNA |
| 18515547 (Nykamp 2008) | Mutant phenotype / localization | Refutes (for larp-1 identity) | Q21992/larp-1 is a DNA-binding TF | LARP-1 has La motif + LARP1 domain, binds RNA in vitro, localizes to germline P bodies, and attenuates Ras-MAPK signaling by controlling pathway mRNA/protein abundance | C. elegans germline/oogenesis | High; primary paper — RNA-binding, not DNA-binding TF |
| Computed provenance (this run) | Computational | Refutes/Competing | Q21992 = csr-1 | 6-mer containment: Q21992↔LARP-1 = 0.971; Q21992↔CSR-1a = 0.000 | Sequence analysis | High; direct calculation |
| UniProt Q17370 (database) | Database | Competing | Source of TF signal | nhr-47 (Nuclear hormone receptor) surfaces on gene:csr-1 text search — a genuine DNA-binding TF but a distinct gene |
C. elegans | Medium; illustrates possible name/paralog confusion |
GO Decision Table (leads — require curator verification)
| Predicted term | Aspect | Applies to CSR-1 Argonaute? | Applies to Q21992/LARP-1? | Recommended action |
|---|---|---|---|---|
| GO:0003700 DNA-binding transcription factor activity | MF | No (PAZ/PIWI RNA slicer) | No (La/LARP1 RNA-binding) | Reject / do not add |
| GO:0019222 regulation of metabolic process | BP | Only as over-broad parent of ncRNA gene silencing | Only as over-broad parent of translation regulation | Do not add — too broad, non-core |
| GO:0009893 positive regulation of metabolic process | BP | No specific support | Weak/broad (positive regulation of translation) | Do not add — too broad, non-core |
| GO:0070551 endoribonuclease cleaving siRNA-paired mRNA | MF | Yes (IDA, WormBase) | n/a | Correct MF for CSR-1 (already annotated) |
| GO:0008266 poly(U) RNA binding | MF | n/a | Yes (IDA, WormBase) | Correct MF for LARP-1 (already annotated) |
GO Curation Implications (leads — require curator verification)
- GO:0003700 (DNA-binding transcription factor activity, MF): DO NOT ADD / REJECT. Refuted for both csr-1 (Argonaute) and Q21992/larp-1 (RNA-binding). No experimental or domain evidence for sequence-specific DNA binding.
- GO:0019222 / GO:0009893 (regulation / positive regulation of metabolic process, BP): treat as NON-CORE and too broad. Do not add on the strength of this prediction. If any regulatory BP is warranted it should be specific and mechanism-appropriate:
- For CSR-1: e.g., regulatory ncRNA-mediated post-transcriptional gene silencing (GO:0035194), siRNA-mediated gene silencing by mRNA destabilization (GO:0090625) — already present with experimental evidence.
- For LARP-1 (if that is the intended protein): positive regulation of translation (GO:0045727) — already present.
- Recommended MF for the actual proteins (informative, not "protein binding"):
- CSR-1: endoribonuclease activity, cleaving siRNA-paired mRNA (GO:0070551); siRNA binding / RNA binding.
- LARP-1: poly(U) RNA binding (GO:0008266); RNA cap binding.
Mechanistic Scope
The predicted MF (DNA-binding TF) posits direct, sequence-specific DNA binding driving transcription. The actual immediate molecular activities are RNA-centric: CSR-1 is a small-RNA-guided endoribonuclease/effector that recognizes target mRNAs by 22G-RNA base-pairing (chromosome segregation, histone-mRNA and germline-mRNA regulation, mRNA surveillance/licensing). LARP-1 is an RNA-binding translational regulator. Any "regulation of metabolic process" is a downstream, category-level description of gene-expression regulation, not a direct DNA-binding transcriptional activity.
Conflicts and Alternatives
- Identifier/accession mismatch (primary conflict): Q21992 (deleted) ≠ csr-1; it is larp-1/R144.7. This is the single most important issue for the curator.
- Winged-helix-like fold artifact: LARP-1 carries a winged-helix-LIKE DNA-binding structural superfamily fold (IPR036388/IPR036390). A predictor keying on that structural signature could spuriously infer "DNA-binding" — but La/winged-helix proteins bind RNA, so this is a false positive.
- Paralog/name confusion: A text search on
gene:csr-1also surfaces nhr-47 (a nuclear hormone receptor and bona fide DNA-binding TF). Frequency bias toward abundant TF/NHR classes plus fuzzy name/xref linking is a plausible origin of the GO:0003700 prediction, but nhr-47 is a separate gene. - No primary literature attributes DNA-binding transcription-factor activity to CSR-1 or LARP-1.
Knowledge Gaps
- Which protein does the prediction actually target? Checked: Q21992 is deleted and maps to larp-1; the symbol says csr-1 (Argonaute). Resolution needed from the curator/model provenance. Matters because the correct-function assignment differs (Argonaute vs La-protein), though the TF verdict is the same. Resolve by confirming the sequence the model was given (compare to H2KZD5/Q27GU1 vs D5MCN2).
- Did BioReason ingest the stale/deleted record? If the model used Q21992's larp-1 sequence but the symbol csr-1, that is a data-hygiene error worth flagging upstream.
- Basis of the TF signal. Not directly observable here; a domain/embedding attribution (winged-helix-like fold vs nhr text co-occurrence) would confirm the false-positive mechanism.
Discriminating Tests
- Sequence-identity check: Align the exact sequence behind the prediction against H2KZD5/Q27GU1 (CSR-1) and D5MCN2 (LARP-1); % identity resolves identity instantly.
- Domain scan (InterProScan/HMMER): Presence of PIWI+PAZ ⇒ Argonaute; La+DM15 ⇒ LARP; absence of homeodomain/zinc-finger/bHLH/NHR DNA-binding domains ⇒ not a TF.
- DNA vs RNA binding assay class: Curated evidence for CSR-1 is IP-small-RNA-seq / mRNA binding; for a TF one would expect ChIP-seq/EMSA on DNA motifs — none exists.
- Compare against
nhr-47(Q17370): confirms the true DNA-binding TF neighbor is a different gene.
Curation Leads
- Action: Reject the BioReason-Pro prediction of GO:0003700 for csr-1/Q21992 as an over-annotation/mis-identification. Do not add GO:0019222 or GO:0009893 (too broad / non-core).
- Flag the accession: Note in the review that Q21992 is a DELETED UniProt entry corresponding to larp-1 (R144.7), not the csr-1 Argonaute locus (F20D12.1; current H2KZD5 / Q27GU1). Recommend the review use a current csr-1 accession.
- Candidate references to verify (exact snippets):
- 19804758 — "the Argonaute CSR-1 … are required for proper chromosome segregation."
- 24178449 — "CSR-1 interacts with small RNAs known as 22G-RNAs … the only essential C. elegans Argonaute out of 24 family members."
- 22863779 — "CSR-1 directly binds to histone mRNA in an ego-1-dependent manner."
- Suggested curator questions: (a) Which sequence did the model score? (b) Is the TF call driven by the LARP winged-helix-like fold or by nhr-name co-occurrence? (c) Should the review record be re-keyed to a live csr-1 accession?
- Suggested experiment (if identity ever in doubt): InterProScan + reciprocal-best-hit orthology to human AGO/PIWI vs LARP1; ChIP-seq would be expected to be negative for DNA-motif occupancy.
Provenance: UniProt REST (Q21992 status, H2KZD5, Q27GU1, D5MCN2), UniParc UPI000007665A cross-references, NCBI eutils esummary (NP_001040868 = "La-related protein 1"), and PubMed primary/review literature. All retrievals executed programmatically in this run.