MOH1 (YBL049W / P38191) — Bioinformatics analysis

Reproducible sequence/domain analysis to ground functional inference for a dark
gene. All numbers below are produced by the scripts in scripts/ from the input
sequence in data/; nothing is hardcoded. Re-run with just all.

Objectives

1. Verify the Yippee/Mis18/Cereblon (CULT) zinc-binding domain and the four
   putative Zn-coordinating cysteines annotated by UniProt (C45, C48, C101, C104).
2. Quantify sequence conservation of MOH1 versus the five human orthologs
   (YPEL1–YPEL5) and test whether the Zn-cysteines are conserved.
3. Assess whether there is any sequence-level evidence for catalytic activity.

Methods

• Input: MOH1 sequence P38191 (data/MOH1.fasta, 138 aa).
• Zinc-ligand scan (scripts/analyze_yippee.py): enumerates all Cys/His and
  finds all C-x(2)-C pairs by regex; reports the candidate Yippee zinc site
  (two CxxC pairs + inter-pair spacer).
• Conservation (scripts/conservation.py): global BLOSUM62 pairwise
  alignment (Biopython PairwiseAligner) of MOH1 against each human YPEL; reports
  percent identity over aligned columns and whether each probed query position is
  identical/similar/gapped in the homolog.
• Orthologs (scripts/fetch_orthologs.py): live download of reviewed human
  YPEL1–5 from UniProt (O60688, Q96QA6, P61236, Q96NS1, P62699).
• Negative control: yeast pyruvate kinase CDC19 (P00549), a non-Yippee
  protein, run through both scripts to confirm they are not hardcoded.

Results

1. Yippee zinc site (from results/moh1_yippee.txt)

length                    138
cysteine_positions        [45, 48, 97, 101, 104, 135, 137]
CxxC_pairs                [(45, 48), (101, 104)]
candidate_zinc_site       C45-x2-C48 ... C101-x2-C104
inter_pair_spacer_residues 52
predicted_zn_ligand_count 4

MOH1 contains exactly the two CxxC motifs that define the Yippee zinc site:
C45-x2-C48 … C101-x2-C104, separated by 52 residues. This is the canonical
Yippee signature (Cys-X₂-Cys-X₅₂-Cys-X₂-Cys) reported in the literature for the
family and matches the four UniProt Zn(2+)-binding features (positions 45, 48,
101, 104) exactly. The additional cysteines (97, 135, 137) do not form the
site.

2. Ortholog conservation (from results/conservation.tsv)

• MOH1 is a bona fide, moderately conserved (~37–45% identity) ortholog of
  the human YPEL family across the Yippee domain.
• All four zinc-binding cysteines are perfectly conserved in every human
  ortholog (= in every cell), strongly supporting a structural Zn(2+) site
  retained from yeast to human.
• Notably, YPEL5 (P62699) is the least identical of the five (37.3%), yet
  it is (a) the PANTHER seed for the IBA "ubiquitin ligase complex" annotation
  and (b) the human gene shown to functionally complement moh1Δ (PMID:28173693).
  So MOH1 is the yeast representative of the whole family rather than a specific
  1:1 ortholog of any single human YPEL.

3. Evidence regarding catalytic activity

The Yippee fold (two four-stranded β-meanders coordinating one Zn²⁺ at the apex,
forming a cradle-shaped pocket) is a structural/binding scaffold, not a
recognized catalytic fold. The sequence contains no canonical catalytic motif
and no EC number is assigned in UniProt/GOA. The conserved cysteines are the
Zn-structural ligands, not a catalytic dyad/triad. There is no sequence-level
evidence for enzymatic activity; the family is best described as a putative
zinc-dependent protein/nucleic-acid interaction module of unknown biochemical
output.

4. Negative control (from results/control_yippee.txt, results/control_conservation.tsv)

Yeast pyruvate kinase CDC19 (P00549, 500 aa) returns
candidate_zinc_site NONE (fewer than two CxxC pairs) and its residues at the
probed positions do not match the YPEL cysteines (mostly gaps). This confirms the
scripts compute results from the actual input rather than emitting a fixed answer.

Conclusions

• Confirmed: MOH1 is a Yippee-family protein with an intact, evolutionarily
  conserved zinc-binding site (C45/C48/C101/C104); the UniProt Zn features and the
  PROSITE Yippee domain call are corroborated independently here.
• Confirmed: MOH1 is a genuine ortholog of the human YPEL1–5 family
  (~37–45% identity), the single yeast member of a family that expanded to five
  paralogs in humans.
• Inconclusive / unknown (correctly so): the analysis provides no
  evidence for a specific molecular function (catalytic or otherwise) beyond
  zinc binding. The biochemical activity and direct partners of MOH1 remain
  undetermined by sequence analysis — consistent with its status as a
  functionally dark gene.

Reproducibility checklist

• [x] No script uses hardcoded inputs or outputs (all read data/*.fasta, write results/*).
• [x] Scripts tested on an independent input (CDC19 / P00549 negative control): Yippee scan correctly returns NO site; conservation probe correctly returns non-matching positions.
• [x] Analyses completed as expected.
• [x] Direct script outputs are in results/.
• [x] Summary includes provenance and justification; uncertainty (catalytic activity, specific MF) explicitly flagged as unresolved.

Tools & data

• Biopython 1.87 (PairwiseAligner, BLOSUM62); Python ≥3.11; requests.
• UniProt REST API (sequences downloaded live).
• Reference sequence: UniProt P38191 (MOH1_YEAST).