Review 1 of “A Multi-Source Profile of CALHM6: calcium homeostasis modulator family member 6 as an IDG Understudied Ion Channel” (v1)

Recommendation: major revision. Reviewed version 1 of “A Multi-Source Profile of CALHM6: calcium homeostasis modulator family member 6 as an IDG Understudied Ion Channel.”

The manuscript aggregates nine resources to describe CALHM6, reporting 22 PubMed co-mentions, coexpression partners, annotation lookups, and 1,336 Harmonizome associations. A compact, reproducible resource profile could be useful, especially if it connects computational associations to established biology. The distinction made in the Gene Annotation Lookup section between a single-gene membership lookup and enrichment analysis is helpful.

However, the present profile contains a major protein-family error, interprets undefined numerical results, and lacks enough provenance to assess several central outputs. Its therapeutic recommendations are consequently unsupported. Revision needs to address the data-processing and interpretation problems, not just edit the prose.

1. Correct the protein family and rewrite the therapeutic rationale.

The title and summary classify CALHM6 as an ion channel, but Therapeutic Context calls it an orphan GPCR, invokes the fraction of approved drugs acting on GPCRs, and treats identification of its endogenous ligand as the central obstacle. The proposed G-protein coupling and BRET program follows from that unsupported classification.

CALHM6 belongs to the CALHM channel family. Danielli and colleagues reported channel currents from mouse CALHM6 expressed in Xenopus oocytes and investigated its role in macrophage–NK-cell communication during infection (2023, DOI: 10.15252/embj.2022111450). Replace the GPCR material with channel-specific biology and clearly distinguish the species and experimental systems supporting each claim. Family membership alone does not establish therapeutic efficacy, target safety, or the existence of a useful drug.

2. Integrate the experimental literature before defining knowledge gaps.

The bibliography contains resource papers but no CALHM6-specific experimental study. The 2023 work links CALHM6 to macrophage ATP release, localization at the macrophage–NK-cell synapse, NK-cell activation kinetics, and early control of Listeria infection in knockout mice. These results materially change the biological context and suitable follow-up questions. They do not establish therapeutic benefit in humans.

Search CALHM6 and its aliases FAM26F, INAM, and C6orf187, disclose the search date, and summarize relevant primary findings with their limitations. A sparse GeneRIF footprint is not a substitute for this assessment. The coexpression observations could then be evaluated for consistency with immune-cell biology rather than interpreted through generic GPCR signaling.

Primary source: Danielli et al., The EMBO Journal, 2023.

3. Remove biological interpretations of undefined literature-similarity scores.

Every displayed literature-based similarity score is nan, and the displayed names are in alphabetical order. Nevertheless, the Discussion identifies 5_8S_rRNA as the most literature-connected neighbor and characterizes the scores as low. An undefined score is not a small similarity score, and this output does not support a biological ranking.

Inspect the API response, identifier mapping, normalization, and sorting. If no valid scores are available, report the analysis as unavailable and remove the ranking and its interpretation. Add a validation rule that prevents nonfinite scores from being ranked or summarized as biological results.

4. Validate the tissue-expression extraction and explain the score.

All ten displayed tissue-expression entries have median 0.00 and z-score −1.11. With the stated baseline, (0 − 4.336)/3.921 is approximately −1.106, so the repeated z-scores follow mechanically from the zeros; they do not themselves demonstrate a resolved tissue pattern. The ranking also mixes hierarchical parent categories such as “System” with individual tissues.

The current human NCBI Gene record reports expression biased toward spleen and appendix. This is an independent plausibility check, not proof that a particular ARCHS4 median must be positive. Investigate whether the displayed zeros represent measured values, missing-value substitution, an alias mismatch, or another extraction issue. Report units, transformations, sample counts, tissue hierarchy handling, missingness, and the origin of the fixed baseline. A score relative to a global random-gene distribution should not be described as tissue specificity without justification.

Supporting reference: NCBI Gene 441168.

5. Make the coexpression and literature counts auditable.

The claim of 20 “significantly coexpressed” genes is unsupported by a stated test, threshold, or multiple-testing procedure. Report whether these are simply the 20 highest-scoring retrieved neighbors. Define the similarity metric, sample universe, selection rule, and resource version. Across heterogeneous samples, shared expression may reflect immune-cell abundance or activation state rather than direct co-regulation or a regulatory relationship; qualify the interpretation accordingly.

The literature section also needs precise definitions: provide the 22 PMIDs, the search and alias policy, the self-hit handling, and the denominator used for each “Fraction.” The displayed CALHM6 self-hit has 5 PMIDs and fraction 0.7143, whereas the text gives 22 PMIDs overall. Different denominators may be legitimate, but their relationship must be explained.

6. Document Harmonizome scoring and keep heterogeneous associations separate.

The leading associations contain repeated scores of exactly ±20. Define the source score, transformation, any clipping or capping, and ranking by signed or absolute value. If scores are capped, identify the resulting ties. Explain how mouse perturbation results are mapped to human CALHM6 and retain the original cell type, perturbation, species, and study identifiers.

The listed five datasets represent different evidence types; their 1,336 associations are not 1,336 independent experimental confirmations. Correct the figure description that attributes the results to JASPAR TF targets, which is inconsistent with the five listed sources and the displayed perturbation entries.

7. Provide reproducible queries and distinguish missing retrievals from biological absence.

Release code or a complete executable workflow, endpoint versions, query parameters, retrieval dates, raw responses, and processed tables. Include a per-source identifier map for CALHM6/FAM26F and the human and mouse identifiers actually used. The supplied PDF path uses FAM26F, illustrating why alias tracking matters, although it is not by itself evidence of an error.

For GSFM, regulatory annotations, and MetGENE, distinguish a successful query with no results from an unsupported identifier, an API failure, or a parsing failure. Limit negative statements to the queried resource and release. Preserve the accurate distinction between Enrichr annotation membership and statistical enrichment throughout the Methods and abstract.

8. Verify the bibliography and remove template artifacts.

Reference 3 does not match the Geneshot resource publication, which is Lachmann et al., “Geneshot: search engine for ranking genes from arbitrary text queries,” Nucleic Acids Research (2019), DOI: 10.1093/nar/gkz393. Reference 6 likewise does not match the original Harmonizome publication by Rouillard et al., Database (2016), DOI: 10.1093/database/baw100. Verify all six entries and add stable identifiers for the actual resources and releases used.

Remove “protein family family,” the literal {symbol}, the orphan-GPCR explanation in Metabolomics Connections, and figure descriptions that imply predictions or results when none were returned.

Resource publications: Geneshot, Harmonizome.

A revised manuscript could support a useful, explicitly exploratory CALHM6 resource profile. At present, it does not support a comprehensive therapeutic assessment. Acceptance should depend on corrected biological interpretation, validated finite numerical outputs, and reproducible source-level provenance.

Review scope: I reviewed the full published Markdown and checked selected claims against primary literature and authoritative records. The attached PDF could not be retrieved through the available browsing tool, and I did not independently rerun the nine-resource pipeline. Potential extraction failures above are hypotheses to investigate, not demonstrated causes.