Enum: EffectTypeEnum

An enumeration of statistical metrics and estimators used to quantify the magnitude and direction of an effect or association between the subject and object of an edge. The numeric result is stored in the companion 'effect size' slot.

Permissible Values

Value	Meaning	Description
regression_coefficient	None	The coefficient of an independent variable in a regression model, representing the expected change in the dependent variable per unit change in the predictor. Commonly reported as a beta coefficient in eQTL, pQTL, mQTL, and GWAS analyses.
log2_fold_change	None	The base-2 logarithm of the ratio of a quantity (e.g. gene expression, protein abundance, metabolite concentration) between two conditions. The standard effect size in differential expression and differential abundance analyses across transcriptomics, proteomics, and metabolomics.
wald_ratio	None	A single-variant Mendelian randomization estimator computed as the ratio of the genetic variant-outcome association to the genetic variant-exposure association.
inverse_variance_weighted	None	A Mendelian randomization estimator that combines per-variant Wald ratios using inverse-variance weighting, assuming all instruments are valid (no horizontal pleiotropy).
mr_egger	None	A Mendelian randomization estimator based on Egger regression that allows for directional horizontal pleiotropy by fitting an intercept term; a non-zero intercept indicates pleiotropic bias.
weighted_median	None	A robust Mendelian randomization estimator that provides a consistent causal estimate when at least 50% of the instrument weight comes from valid variants.
standardized_mean_difference	None	The difference between two group means divided by a pooled standard deviation, used when the predictor is categorical and the outcome is continuous.
cohens_d	None	A standardized mean difference using the square root of the average of the two group variances as the denominator, with a small-sample correction for n < 50.
hedges_g	None	A standardized mean difference with a Gamma-function correction on the pooled standard deviation to reduce small-sample bias.
glasss_delta	None	A standardized mean difference using only the control group's standard deviation as the denominator, preferred when group variances differ substantially.
strictly_standardized_mean_difference	None	The ratio of the mean difference to the standard deviation of the difference between two groups (SSMD), widely used in high-content screening for hit selection and quality control.
correlation_coefficient	None	A normalized measure of association between two variables, ranging from -1 to +1, computed as covariance divided by the product of standard deviations.
pearsons_r	None	A parametric correlation coefficient for two continuous, normally distributed variables with a linear relationship.
spearmans_rho	None	A nonparametric rank correlation coefficient measuring monotonic dependence between two variables.
kendalls_tau	None	A rank correlation coefficient based on concordant and discordant pairs, suitable for ordinal data.
polychoric_correlation	None	A correlation coefficient estimating the association between two latent continuous variables from observed ordinal data.
matthews_correlation_coefficient	None	A correlation-based measure of binary classification quality (also known as the phi coefficient), ranging from -1 to +1.
goodman_kruskal_gamma	None	A rank correlation measure for ordinal or continuous variables based on the difference between concordant and discordant pairs, excluding ties.
r2_linkage_disequilibrium	None	A squared correlation coefficient over two dichotomous variables used as a measure of linkage disequilibrium.
odds_ratio	None	The ratio of the odds of an outcome in an exposed group to the odds in an unexposed group, measuring association strength between two binary variables.
relative_risk	None	The ratio of event probability in an exposed group to event probability in an unexposed group (also called risk ratio).
hazard_ratio	None	The ratio of hazard rates between two groups over time, commonly reported in survival analysis and time-to-event omics studies.
eta_squared	None	The proportion of total variance in the dependent variable attributable to a predictor in the sample (a biased estimator analogous to R-squared).
omega_squared	None	A less biased estimator of the proportion of variance explained by a predictor, providing a better population-level estimate than eta-squared.
root_mean_square_standardized_effect	None	The square root of the average of squared standardized effects in an analysis of variance context (RMSSE, Psi).

Slots constrained by this enum

Name
effect_type

LinkML Source

name: EffectTypeEnum
description: An enumeration of statistical metrics and estimators used to quantify
  the magnitude and direction of an effect or association between the subject and
  object of an edge. The numeric result is stored in the companion 'effect size' slot.
from_schema: https://w3id.org/biolink/vocab/
permissible_values:
  regression_coefficient:
    text: regression_coefficient
    description: The coefficient of an independent variable in a regression model,
      representing the expected change in the dependent variable per unit change in
      the predictor. Commonly reported as a beta coefficient in eQTL, pQTL, mQTL,
      and GWAS analyses.
    close_mappings:
    - STATO:0000565
  log2_fold_change:
    text: log2_fold_change
    description: The base-2 logarithm of the ratio of a quantity (e.g. gene expression,
      protein abundance, metabolite concentration) between two conditions. The standard
      effect size in differential expression and differential abundance analyses across
      transcriptomics, proteomics, and metabolomics.
    close_mappings:
    - STATO:0000169
  wald_ratio:
    text: wald_ratio
    description: A single-variant Mendelian randomization estimator computed as the
      ratio of the genetic variant-outcome association to the genetic variant-exposure
      association.
  inverse_variance_weighted:
    text: inverse_variance_weighted
    description: A Mendelian randomization estimator that combines per-variant Wald
      ratios using inverse-variance weighting, assuming all instruments are valid
      (no horizontal pleiotropy).
  mr_egger:
    text: mr_egger
    description: A Mendelian randomization estimator based on Egger regression that
      allows for directional horizontal pleiotropy by fitting an intercept term; a
      non-zero intercept indicates pleiotropic bias.
  weighted_median:
    text: weighted_median
    description: A robust Mendelian randomization estimator that provides a consistent
      causal estimate when at least 50% of the instrument weight comes from valid
      variants.
  standardized_mean_difference:
    text: standardized_mean_difference
    description: The difference between two group means divided by a pooled standard
      deviation, used when the predictor is categorical and the outcome is continuous.
    close_mappings:
    - STATO:0000100
  cohens_d:
    text: cohens_d
    description: A standardized mean difference using the square root of the average
      of the two group variances as the denominator, with a small-sample correction
      for n < 50.
    is_a: standardized_mean_difference
    close_mappings:
    - STATO:0000618
  hedges_g:
    text: hedges_g
    description: A standardized mean difference with a Gamma-function correction on
      the pooled standard deviation to reduce small-sample bias.
    is_a: standardized_mean_difference
    close_mappings:
    - STATO:0000319
  glasss_delta:
    text: glasss_delta
    description: A standardized mean difference using only the control group's standard
      deviation as the denominator, preferred when group variances differ substantially.
    is_a: standardized_mean_difference
    close_mappings:
    - STATO:0000320
  strictly_standardized_mean_difference:
    text: strictly_standardized_mean_difference
    description: The ratio of the mean difference to the standard deviation of the
      difference between two groups (SSMD), widely used in high-content screening
      for hit selection and quality control.
    is_a: standardized_mean_difference
    close_mappings:
    - STATO:0000135
  correlation_coefficient:
    text: correlation_coefficient
    description: A normalized measure of association between two variables, ranging
      from -1 to +1, computed as covariance divided by the product of standard deviations.
    close_mappings:
    - STATO:0000142
  pearsons_r:
    text: pearsons_r
    description: A parametric correlation coefficient for two continuous, normally
      distributed variables with a linear relationship.
    is_a: correlation_coefficient
    close_mappings:
    - STATO:0000280
  spearmans_rho:
    text: spearmans_rho
    description: A nonparametric rank correlation coefficient measuring monotonic
      dependence between two variables.
    is_a: correlation_coefficient
    close_mappings:
    - STATO:0000201
  kendalls_tau:
    text: kendalls_tau
    description: A rank correlation coefficient based on concordant and discordant
      pairs, suitable for ordinal data.
    is_a: correlation_coefficient
    close_mappings:
    - STATO:0000240
  polychoric_correlation:
    text: polychoric_correlation
    description: A correlation coefficient estimating the association between two
      latent continuous variables from observed ordinal data.
    is_a: correlation_coefficient
    close_mappings:
    - STATO:0000269
  matthews_correlation_coefficient:
    text: matthews_correlation_coefficient
    description: A correlation-based measure of binary classification quality (also
      known as the phi coefficient), ranging from -1 to +1.
    is_a: correlation_coefficient
    close_mappings:
    - STATO:0000524
  goodman_kruskal_gamma:
    text: goodman_kruskal_gamma
    description: A rank correlation measure for ordinal or continuous variables based
      on the difference between concordant and discordant pairs, excluding ties.
    is_a: correlation_coefficient
    close_mappings:
    - STATO:0000612
  r2_linkage_disequilibrium:
    text: r2_linkage_disequilibrium
    description: A squared correlation coefficient over two dichotomous variables
      used as a measure of linkage disequilibrium.
    is_a: correlation_coefficient
    close_mappings:
    - STATO:0000123
  odds_ratio:
    text: odds_ratio
    description: The ratio of the odds of an outcome in an exposed group to the odds
      in an unexposed group, measuring association strength between two binary variables.
    close_mappings:
    - STATO:0000182
  relative_risk:
    text: relative_risk
    description: The ratio of event probability in an exposed group to event probability
      in an unexposed group (also called risk ratio).
    close_mappings:
    - STATO:0000245
  hazard_ratio:
    text: hazard_ratio
    description: The ratio of hazard rates between two groups over time, commonly
      reported in survival analysis and time-to-event omics studies.
    close_mappings:
    - STATO:0000677
  eta_squared:
    text: eta_squared
    description: The proportion of total variance in the dependent variable attributable
      to a predictor in the sample (a biased estimator analogous to R-squared).
    close_mappings:
    - STATO:0000317
  omega_squared:
    text: omega_squared
    description: A less biased estimator of the proportion of variance explained by
      a predictor, providing a better population-level estimate than eta-squared.
    close_mappings:
    - STATO:0000318
  root_mean_square_standardized_effect:
    text: root_mean_square_standardized_effect
    description: The square root of the average of squared standardized effects in
      an analysis of variance context (RMSSE, Psi).
    close_mappings:
    - STATO:0000316