CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

bio-causal-genomics-mediation-analysis

Decompose genetic effects into direct and indirect paths through mediating variables using the mediation R package. Tests whether gene expression, methylation, or other molecular phenotypes mediate the effect of genetic variants on disease. Use when testing whether a molecular...

Quality

73%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./scientific-skills/Data Analysis/bio-causal-genomics-mediation-analysis/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

57%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-organized, code-first skill whose R examples are concrete and largely copy-paste ready across the main genomic mediation scenarios. Its weaknesses are template boilerplate that doesn't fit an R skill, a batch mediation loop lacking validation/error handling, and a monolithic single-file structure with no progressive disclosure of advanced topics.

Suggestions

Remove or rewrite the boilerplate sections for this R skill: the Version Compatibility section cites Python exceptions (ImportError/AttributeError/TypeError), and the generic Input Validation footer adds no domain-specific guidance.

Add per-gene error handling and validation to the batch eQTL mediation loop (e.g. tryCatch around each run_eqtl_mediation call, or a convergence/model-fit check before mediate()) so one failing gene doesn't abort the entire batch.

Split advanced topics (HIMA high-dimensional mediation, multi-omics chains, visualization) into one-level-deep reference files under references/ and signal them from SKILL.md, keeping the overview lean.

DimensionReasoningScore

Conciseness

The code sections are lean and the inline comments earn their place, but there is noticeable padding: the Version Compatibility boilerplate references Python exceptions ("If code throws ImportError, AttributeError, or TypeError") in an R-only skill, the generic 'Input Validation' footer adds no domain information, and the Framework section restates mediation concepts Claude already knows. This matches anchor 3 ('mostly efficient but includes some unnecessary explanation') better than anchor 4, where such boilerplate would be absent.

3 / 5

Actionability

Concrete, mostly executable R code covers the common cases: basic mediate() with bootstrap, extracting ACME/ADE/total/proportion-mediated, a reusable run_eqtl_mediation() function with FDR adjustment, multi-omics chains, HIMA, medsens() sensitivity, and ggplot visualization. Minor gaps keep it below anchor 5: examples depend on an undefined `dat` frame and undefined `mediator_cols`/`covariate_cols`, and the two plot functions are defined but never invoked.

4 / 5

Workflow Clarity

The basic workflow is well sequenced (explicit Step 1/2/3 comments for mediator model, outcome model, mediate()) and the version-check section provides an error-adaptation loop, but the eQTL mediation is a batch operation (lapply over many genes) with no per-item validation or error handling — one failing model aborts the whole run — and no model-fit checks precede mediate(). Per the rubric cap, a batch workflow without validation cannot score above 3.

3 / 5

Progressive Disclosure

There are no bundle files at all (no references/, scripts/, or assets/), and the entire ~270-line body — including self-contained advanced topics (HDMA/HIMA, multi-omics chains, visualization) — is inlined in SKILL.md. Section headers are clear and well organized, but content that would merit splitting into one-level-deep reference files is all inline with no navigation to separate materials, matching anchor 3.

3 / 5

Total

13

/

20

Passed

Description

78%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, domain-specific description with concrete actions, good natural trigger terms, and a distinct niche. Its main defect is that it is truncated mid-sentence ("Use when testing whether a molecular..."), which cuts off the explicit trigger guidance and drops completeness and trigger-term coverage below excellent.

Suggestions

Complete the truncated description — the 'Use when testing whether a molecular...' clause is cut off mid-phrase at ~290 characters, losing the explicit trigger scenario.

Add natural trigger synonyms users would actually say, e.g. 'mediation analysis', 'indirect effect', 'ACME/ADE', 'eQTL', or 'proportion mediated'.

DimensionReasoningScore

Specificity

The description lists several concrete actions — "Decompose genetic effects into direct and indirect paths through mediating variables using the mediation R package" and "Tests whether gene expression, methylation, or other molecular phenotypes mediate the effect of genetic variants on disease" — naming the tool, the mediator types, and the outcome. It falls short of anchor 5 because it omits capabilities the body actually covers (eQTL batch mediation, multi-omics chains, high-dimensional HIMA, sensitivity analysis), leaving minor gaps in coverage.

4 / 5

Completeness

The "what" is clear (decompose genetic effects into direct/indirect paths via the mediation R package) and a "Use when" clause is present ("Use when testing whether a molecular..."), placing it above anchor 3 where 'when' is only weakly implied. It does not reach anchor 5 because the 'when' clause is literally truncated mid-phrase in the file, so the explicit concrete trigger scenario is incomplete.

4 / 5

Trigger Term Quality

Natural terms a genomics user would say are present: "mediate", "gene expression", "methylation", "genetic variants", "disease", "mediation". It is not a 5 because common variations like "mediation analysis", "indirect effect", "ACME", "eQTL", or "SNP" are absent, and the truncated tail ("Use when testing whether a molecular...") cuts off whatever trigger phrases followed.

4 / 5

Distinctiveness Conflict Risk

"Decompose genetic effects into direct and indirect paths through mediating variables using the mediation R package" carves out a clear niche with distinct triggers (mediation via expression/methylation mediators). Adjacent skills like Mendelian randomization or GWAS association testing would trigger on different phrasing, so conflict risk is minimal.

5 / 5

Total

17

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

aipoch/medical-research-skills

Path

scientific-skills/Data Analysis/bio-causal-genomics-mediation-analysis/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.