Rv3599c Still unknown · low auto-curated · to review
H37Rv Rv3599c · MTBC0 ·
27 aa ·
4042952–4043035 H37Rv
(-) ·
RefSeq
Genomic neighbourhood (genome browser)
Open in full genome browser →This gene (outlined) in its genomic context; arrows are neighbouring genes coloured by verdict. Click any gene to navigate. Pan and zoom in the full browser.
Annotation: from legacy to revised
| Legacy (H37Rv / Mycobrowser) | Hypothetical short protein |
|---|---|
| MTBC0 PGAP re-annotation | — |
| Functional category (TubercuList) | unknown |
Curation note: Added P16.5d: Mycobrowser-annotated gene absent from RefSeq NC_000962.3 (the atlas' original 3906-CDS reference). Foundation record from the Mycobrowser release + translated H37Rv sequence; heavy enrichment layers (structure, orthology, conservation, interaction) pending.
Auto-curated: this verdict and function were generated by rules from PGAP + Pfam + Foldseek and have not been hand-reviewed (flagged for human review).
In the literature (TB corpus sweep) no title/abstract mention
No TB publication mentions this locus tag in its title or abstract (sweep of a ~330k-abstract PubMed TB corpus). A full-text search does however find 1 candidate paper(s) whose full text is not open access and could not be read here, so this page does not claim that the gene was never studied: the mention may be incidental or substantive. The gene remains dark on function.
Second-source check (P43.6): tbmonitor finds 0 paper(s) naming this locus tag, the PubMed sweep 0 — no mention in either source. Neither source is complete: across the 210 fiches whose source verdict is dark, 3 are seen only by tbmonitor and 4 only by PubMed, which is why a 'no mention in the literature' statement is never made on a single source here. (The site's own counter reports 209: Rv2438A carries a curated function and is recalled to family_assigned at ingest by the verdict/function coherence guard, so it leaves the dark tally without leaving this cross-check.) A count above zero says the gene has been NAMED in a paper, not that its molecular function is known: Rv2660c carries 38 papers as the latency antigen of the multistage vaccine H56 (Ag85B + ESAT-6 + Rv2660c, Aagaard et al., Nat Med 2011, PMID 21258338) and is still dark on function. Counts are shown side by side and never merged.
Scope of the sweep: What was actually swept: the H37Rv locus tag and the ortholog locus tags (M. bovis Mb…, M. marinum MMAR_…, M. smegmatis MSMEG_…, M. leprae ML…, M. abscessus MAB_…), plus the gene name where one exists. NOT swept: protein and enzyme names, including family shorthands such as Ldt(Mt4) or VapC28, and their orthographic variants. An absent locus-tag mention is therefore evidence about the TAG, and only indirect evidence about the protein (P18.14).
An antigen status or a virulence phenotype is NOT a molecular function: the verdict stays unchanged. This layer adds the missing context, it does not requalify the gene. This distinguishes a gene that is dark because nobody has looked from one that is dark despite having been studied. Source: PubMed (whole), MULTI-ALIAS sweep of title+abstract: the H37Rv locus tag AND every ortholog identifier (M. bovis Mb…, M. marinum MMAR_…, M. smegmatis MSMEG_…, M. leprae ML…, M. abscessus MAB_…), each hit VERIFIED against the abstract text (word-boundary regex). phase73, 2026-07-13.
Why this gene is still unknown (classification, not failure) MICRO-ORF
The detectors cannot speak here. WALL OF MEASURABILITY. Below ~65 aa a protein cannot carry a full Pfam-A domain, profile-profile search has too little sequence to build a deep profile, and a TM-score around 0.6 against any small fold is ordinary rather than significant. Pocket detection is blind too: on PROVEN enzymes, only 1 of 26 under 200 aa reaches the P2Rank confidence threshold. The absence of a functional call here measures the detectors, not the protein.
| Detector state | length 27 aa · Pfam domain: none · confident AlphaFold model: yes · significant structural hits: 0 · pocket detection: blind (protein below the detector's power) · MS: not detected |
|---|---|
| Existence vs function | Not MS-detected: neither existence nor function is settled here. A proteomics negative is weak evidence for short, hydrophobic or low-abundance proteins. |
| What would settle it | Bench, or a detector built for short proteins. Existence is often provable (MS); function is not inferable in silico with today's tools. |
Read with care. A statement about what can be MEASURED on this gene today, not a claim that it has no function, and revisable: a gene enters or leaves this class as the layers change. It stays counted among the dark. phase117_hard_residue.py (P8-résidu-dur), 2026-08-10.
Binding-pocket screen (P2Rank, geometric prediction) detector blind at this length
| Pockets found | 0 |
|---|---|
| Model length screened | 27 aa |
Read with care. This protein (27 aa) is below the size where this detector has meaningful power: on proven enzymes, only 3.8% (1/26) under 200 aa reach this layer's P2Rank confidence threshold of best_probability >= 0.70, versus 60.5% (75/124) above it (P16.3b calibration, negative control EsxA/EsxB-scale panel). Both figures are recall AT 0.70, which is also the best separation this detector reaches on that panel (Youden J = 0.416) and whose false-positive rate above 200 aa is 18.9% (81.1% specificity); the looser 0.50 that appears elsewhere in this repository would raise recall above 200 aa to 74.2% and drop specificity to 64.9%. A negative or weak pocket result here should NOT be read as evidence against a ligand-binding role -- the test essentially has no power at this length, not that the protein lacks a site. The same holds in the other direction, and that half was missing here: a STRONG score at this length is not a result either. The same calibration panel does carry that other arm below 200 aa, and the two do not separate: the 3.8% above is matched by 6.6% (5/76) of proteins annotated as non-catalytic clearing 0.70 (Fisher p = 1.00; 15.4% against 11.8% at 0.50, p = 0.73). At this length the score therefore carries no measured information in either direction -- which is a stronger statement than untested power, and the reason a strong score here is not a result. P16.3e, calibration phase72b_pocket_calibration.py, 2026-08-03.
Legacy record & comparison (Mycobrowser)
| Mycobrowser function | Unknown |
|---|
The legacy Mycobrowser record is shown for verification. Mycobrowser is no longer maintained; its EC numbers predate recent nomenclature revisions, so a class change usually reflects re-numbering, not a conflict.
Curated reference (UniProt)
| UniProt |
O06283
TrEMBL · unreviewed
|
|---|---|
| UniProt name | Hypothetical short protein |
Conservation & selection (intra-MTBC, 142 601 strains, snapshot of 2026-09-29)
| pN/pS | n/a |
|---|---|
| Polymorphic sites (≥ 0.1% of strains) | 0 synonymous, 1 missense, 0 nonsense, 0 frameshift |
pN/pS from segregating SNPs (singletons removed) normalised by possible sites. Low pN/pS = purifying selection (a strong signal that a "hypothetical" is a real, constrained gene). A high pN/pS is ambiguous: relaxed constraint or positive selection (drug resistance, antigenic variation) inflate it; e.g. rpoB/katG/pncA score high here for resistance, not loss of function. A clonal disruption (one allele over a clade) suggests lineage pseudogenisation; a convergent one (many independent alleles) is typical of resistance loss-of-function.
Essentiality (transposon mutagenesis)
| DeJesus 2017 call | Uncertain · uncertain |
|---|---|
| What the call means | uncertain (short or TA-poor ORF): no call possible |
| TA sites (Himar1) |
2 in the ORF —
0 in the essential state,
0 growth-defect,
2 non-essential,
0 growth-advantage.
Saturation 0.500,
mean read count 361.
A region of the protein devoid of TA sites is invisible to this assay: nothing can be
inferred about it, in either direction.
All 2 sites lie in this ORF alone (no overlapping CDS shares any), so the call is attributable to this gene. |
| Caveat | Statistically thin call: only 2 TA (Himar1) sites in the whole ORF (atlas median 13; genes under 300 nt typically have very few). A DeJesus 2017 call built on so few independent observations is less robust than the same call on a longer gene, in either direction. Cross-check against the CRISPRi vulnerability index (independent of TA-site density) and, if this gene overlaps a neighbour (see Genomic-neighbour overlap section below), verify how many of its TA sites actually fall inside its own ORF. (P20.3) |
Genome-wide Himar1 transposon essentiality in H37Rv (DeJesus 2017). An essential call (ES/ESD/GD) is strong, independent evidence that a "hypothetical" locus encodes a functional, selectively required gene — orthogonal to intra-species conservation.
Proteomics (mass spectrometry) not detected
Not detected in the mass-spectrometry datasets integrated by PaxDb. This is not evidence of absence, and this atlas can say which bias dominates, because it measured it on its own 482 undetected proteins. Size is the main factor: undetected proteins have a median length of 145 aa against 295 aa for detected ones, and 25.3 % of them are shorter than 100 aa against 6.0 %. Predicted localisation matters far less than usually assumed: detection rates are 89.3 % for globular, 84.9 % for transmembrane, 88.0 % for lipoproteins and 77.8 % for signal-peptide proteins — so a short protein is under-sampled whatever its predicted compartment, and hydrophobicity alone is a weak excuse. Condition-specific expression and few tryptic peptides (PE/PPE repeats) add to this.
Predicted localisation (DeepTMHMM + lipobox) signal peptide
| Prediction | predicted secreted protein (signal peptide) |
|---|---|
| DeepTMHMM class | SP |
Transmembrane topology and signal peptide from DeepTMHMM (deep-learning reference predictor); lipoproteins from a (myco)bacterial lipobox motif. A sequence-based prediction of subcellular context.
Physico-chemical properties (computed, ProtParam)
| Length | 27 aa |
|---|---|
| Molecular weight | 2.9 kDa |
| Theoretical pI | 6.52 |
| GRAVY | -0.652 (hydrophilic) |
| Aliphatic index | 68.9 |
| Aromaticity | 0.0 |
| Instability index | 93.8 (unstable) |
Computed from the ancestral MTBC0 sequence with the ExPASy ProtParam method (Biopython). Descriptive biophysical context: a positive GRAVY flags a hydrophobic (often membrane) protein, a high instability index (>40) predicts a short in-vitro half-life, an extreme pI hints at compartment or binding partner.
Domains (Pfam, hmmscan --cut_ga)
No Pfam-A domain above the gathering threshold (hmmscan --cut_ga scan completed).
3D structure (interactive — Mol*/pdbe-molstar)
No experimental structure is deposited for this gene: showing the AlphaFold DB predicted model (mean pLDDT 77.0, confident), coloured by per-residue confidence (dark blue = very high, orange/red = low — treat low-confidence regions as unreliable, especially termini and loops). Drag to rotate, scroll to zoom.
Transcriptional regulation (signed TRN: ChIP-seq + TFOE)
| Regulated by (6 TF) |
Rv0135c (represses) · Rv0324 (activates) · Rv1353c (represses) · cmtR (activates) · Rv2011c (represses) · Rv2642 (represses)
|
|---|
Regulatory edges from the ISB signed transcriptional regulatory network (TF ChIP-seq binding, Minch 2015 + TF-overexpression response, Rustad 2014). An edge is regulatory evidence (binding and/or expression change), not necessarily direct. For a "hypothetical", membership in a known regulon (e.g. DosR dormancy, PhoP virulence) is a physiological-context lead, not a function.
Condition: Condition-blind by construction. These edges come from transcription-factor OVEREXPRESSION in a single laboratory condition, so a listed regulator is not a claim that it governs this gene in every circumstance. Measured counter-examples exist within this atlas (see Rv2516c, where the DosR regulon collapses under moxifloxacin while the gene itself rises). Read a regulator as 'capable of shifting this gene', and look for the condition before building a hypothesis on it.
Functional interaction network (STRING v12)
This gene was queried against the STRING v12 network of M. tuberculosis H37Rv and has no partner at or above medium confidence (combined score ≥ 400). This is a verified absence, not a gap in the pipeline: an isolated node at this threshold is itself informative, and frequently the case for short or lineage-restricted proteins.
Sources
- Ancestral sequence & coordinates: Harrison LB et al. (2024), An imputed ancestral reference genome for the MTBC, doi:10.1101/2023.09.07.556366
- Product annotation: NCBI PGAP on MTBC0; legacy from H37Rv NC_000962.3 (RefSeq )
- Domains: Pfam-A via hmmscan --cut_ga — none above threshold
- Sequence-level signal: ESM Atlas (EvolutionaryScale × BioHub) — exploratory
- Curated reference: UniProt O06283 (TrEMBL, unreviewed)
- Intra-MTBC selection: pN/pS and disruption from SPDI variants of 142 601 MTBC strains (this work, local collection vs H37Rv NC_000962.3)
- Essentiality: genome-wide transposon mutagenesis in H37Rv — DeJesus et al. 2017 (mBio, doi:10.1128/mBio.02133-16, CC BY)
- Functional category: TubercuList scheme (Cole et al. 1998, doi:10.1038/31159), via Mycobrowser (Kapopoulou et al. 2011, doi:10.1016/j.tube.2010.09.006)
- Transcriptional regulation: ISB signed TRN — TF ChIP-seq (Minch et al. 2015, doi:10.1038/ncomms6829) + TF overexpression (Rustad et al. 2014, doi:10.1186/gb-2014-15-11-502)
- Physico-chemical properties: ExPASy ProtParam method via Biopython (Gasteiger et al. 2005), computed from the MTBC0 sequence
- Predicted localisation: DeepTMHMM (Hallgren et al. 2022, doi:10.1101/2022.04.08.487609) for transmembrane topology and signal peptide
- Primary literature: none located yet; annotation rests on the domain/homology sources above.
Ancestral MTBC0 protein sequence
>|Rv3599c|Rv3599c MPASSLGTGSPAADRLDATHERRREVI
Spot an error? Suggest an improvement
Found a mistake, a missing reference, or have a better functional hypothesis for Rv3599c? Report it below — the message is emailed to the maintainer with this gene's details, reviewed by hand (no automatic edits).
Prefer email directly? ✉ Email a correction about Rv3599c