Section by section
What to enter in each part of the form.
1 · Method identity
- Give the method's full name, an acronym if it has one, and the version researchers should use.
- Write the short description for someone outside your field: what the method does and for what kind of data.
- Your name and contact email are needed so reviewers can reach you. Your ORCID is optional but helps credit you correctly.
2 · Scientific purposeSELECTION-CRITICAL
- Write the research questions your method answers, phrased the way a researcher would ask them. Add several.
- statLens matches these questions against what users ask, so concrete questions work better than general ones.
- Tick every analysis task the method supports.
3 · Required inputs
- List each input the method needs: its name, type, whether it is required, and the structure you expect.
- statLens convention: omics tables are samples in rows and features in columns; metadata is samples in rows and study variables in columns.
4 · Supported data characteristicsSELECTION-CRITICAL
- Tick only the data characteristics the method actually handles.
- These are hard filters: if a user's data has a characteristic you did not tick, your method will not be suggested for it. Over-ticking leads to misuse.
5 · Study designSELECTION-CRITICAL
- Tick the study designs the method is appropriate for. Like data characteristics, these are hard filters.
6 · Statistical specificationSELECTION-CRITICAL
- Describe the model: family, distribution, link, the estimand and null hypothesis, and how multiple testing is handled.
- Write the model formulation in LaTeX, for example \log E(Y_{ij}) = x_{ij}^\top \beta + b_i.
7 · Assumptions and applicability
- List the assumptions, when the method performs well, and when it should not be used, one item per line.
- Be specific in “Should not be used when”. It prevents misuse more effectively than a caveat in a paper.
8 · Method comparisons
- Link your method to related ones: alternatives, methods it extends, methods it should follow, and what it was validated against.
- Add a note on how the two differ in practice. These links let statLens suggest sequences of methods, not just single choices.
9 · Outputs
- Tick the standard result fields your method's output provides, such as estimate, p_value and adjusted_p_value.
- List any method-specific output fields and the diagnostic plots it produces.
10 · Software implementation
- Say where the software is distributed (CRAN, Bioconductor, PyPI, GitHub…), the package name, licence, repository URL and dependencies.
- Add a short example call that runs the method.
11 · Citation
- Give the DOI and details of the paper that introduces the method. Users of your method are asked to cite it.
12 · Validation and examples
- Link to a small example dataset and the output your method produces on it. Simulated data is fine.
- A reproducible example is required before a method can be verified. For now these are links; direct file upload is coming.
13 · Submission
- Agree to the contributor licence terms, then submit. You can save a draft at any point.