compute.metrics: Performance of Gene Identification
In estfernan/boost: Bayesian Modeling of Spatial Transcriptomics Data

compute.metrics

R Documentation

Performance of Gene Identification

Description

Calculate the performance of spatially variable (SV) gene identification on simulated data.

Usage

compute.metrics(
  predictor,
  truth,
  predictor.type = c("BF", "p-value"),
  threshold = NULL
)

Arguments

`predictor`	A numeric vector of length n that denotes the p-values or Bayes factors (BFs).
`truth`	A logical vector of length n that represents the ground truth corresponding to the predictor.
`predictor.type`	A character string that specifies whether p-values of Bayes factors (BFs) were provided.
`threshold`	A numeric value that specifies the cutoff for defining SV genes.

Value

A list object that contains six performance metrics (Sensitivity, Specificity, F1_score, FDR, AUC, and MCC).

References

Li, X., Wang, X., & Xiao, G. (2019). A comparative study of rank aggregation methods for partial and top ranked lists in genomic applications. Briefings in bioinformatics, 20(1), 178–189. https://doi.org/10.1093/bib/bbx101.

Robin, X., Turck, N., Hainard, A. et al. pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics 12, 77 (2011). https://doi.org/10.1186/1471-2105-12-77.

estfernan/boost documentation built on June 24, 2022, 12:20 a.m.