Input and output formats

Phenotype dataframe

Rows represent samples. Columns contain outcomes and covariates.

Part	Requirement
Index	Must match the predictor dataframe index exactly.
Outcomes	Numeric columns named in the `outcomes` argument.
Covariates	Optional numeric columns named in the `covariates` argument.
Missing values	Allowed. Each regression uses its own complete-case subset.

sample_id  trait_a  trait_b  age  sex
S001       1.2      0        50   1
S002       2.0      1        61   0
S003       1.5      0        47   1

Predictor dataframe

Rows use the same sample index as the phenotype dataframe. Each column is tested independently against every selected outcome.

Predictor type	Example
Genotype dosage	0, 1, and 2 effect-allele copies.
Phenotype	A biomarker or derived measurement.
Score	A polygenic score or model prediction.

PheWAS result dataframe

Column	Description
`outcome`	Outcome phenotype name.
`predictor`	Predictor name.
`sample_count`	Complete-case sample count.
`total_sample_count`	Total aligned input sample count.
`negative_log10_pvalue`	Negative base-10 logarithm of the regression p-value.
`pvalue`	Regression p-value.
`beta`	Predictor coefficient.
`standard_error`	Coefficient standard error.

Phenotype metadata

Metadata provides labels and plotting groups. It may be a dataframe or a CSV/TSV path.

Column	Required	Description
`outcome`	Yes	Identifier matching the PheWAS result.
`category`	Yes	Group used on Manhattan and enrichment plots.
`description`	Optional	Readable label used for annotations.

MR summary statistics

Column	Required	Description
`variant_id`	Yes	Variant identifier.
`chromosome`	Yes	Chromosome. X and Y are accepted.
`effect_allele`	Yes	Allele associated with `beta`.
`other_allele`	Recommended	Other allele used during harmonization.
`beta`	Yes	Effect estimate.
`standard_error`	Yes	Effect standard error.
`pvalue`	Yes	Association p-value.
`sample_size`	Optional	Sample size used to resolve duplicate variant rows.

variant_id  chromosome  effect_allele  other_allele  beta   standard_error  pvalue
rs100001    1           A              G             0.12   0.03            2e-8
rs100002    2           C              T            -0.08   0.02            4e-9

MR result dataframe

Column	Description
`method`	MR estimator name.
`exposure`	Exposure label supplied to the analysis.
`outcome`	Outcome label supplied to the analysis.
`pvalue`	Estimator p-value.
`beta`	Estimated exposure-outcome effect.
`standard_error`	Estimator standard error.
`variant_count`	Number of variants used by the estimator.

Gene interval file

annotate_genes accepts a tab-separated interval table. The variant and gene coordinates must use the same genome assembly.

Column	Description
`chromosome`	Chromosome, such as chr1.
`start`	Gene coding start coordinate.
`end`	Gene coding end coordinate.
`gene`	Gene symbol.

Rename source columns to this schema before calling PYPE. Choose GRCh37/hg19 or GRCh38 to match the variant coordinates.

Plot outputs

Plot functions write PNG, SVG, PDF, or other Matplotlib-supported image formats based on the output suffix. Manhattan and volcano calls also write tab-separated significant-result files beside the image.