predict_snpnet: Predict from the Fitted Object or File
In rivas-lab/snpnet: snpnet: Efficient Lasso Solver for Large SNP Data

Predict from the Fitted Object or File

predict_snpnet(fit = NULL, saved_path = NULL, new_genotype_file, new_phenotype_file,
  phenotype, gcount_path = NULL, meta_dir = NULL, meta_suffix = ".rda", covariate_names = NULL,
  split_col = NULL, split_name = NULL, idx = NULL, family = NULL, snpnet_prefix = "output_iter_",
  snpnet_suffix = ".RData", snpnet_subdir = "results", configs = list(zstdcat.path = "zstdcat",
  zcat.path='zcat'))

`fit`	Fitted object returned from the snpnet function. If not specified, 'saved_path' has to be provided.
`saved_path`	Path to the file that saves the fit object. The full path is constructed as $saved_path/$snpnet_subdir/$snpnet_prefixITER$snpnet_suffix, where ITER will be the maximum index found in the snpnet subdirectory. If not specified, 'fit' has to be provided.
`new_genotype_file`	Path to the new suite of genotype files. new_genotype_file.pgen, psam, pvar.zst. must exist.
`new_phenotype_file`	Path to the phenotype. The header must include FID, IID. Used for extracting covaraites and computing metrics.
`phenotype`	Name of the phenotype for which the fit was computed.
`gcount_path`	Path to the saved gcount file on which the meta statistics can be computed. Only if 'saved_path' is specified.
`meta_dir`	(Depreciated) Path to the saved meta statistics object. The full path is constructed as $meta_dir/$STAT$meta_suffix, where such files should exist for STAT = pnas, means and optionally sds. Only if 'saved_path' is specified.
`meta_suffix`	(Depreciated) Extension suffix of the meta statistics files. Only if 'saved_path' is specified.
`covariate_names`	Character vector of the names of the adjustment covariates.
`split_col`	Name of the split column. If NULL, all samples will be used.
`split_name`	Vector of split labels where prediction is to be made. Should be a combination of "train", "val", "test".
`idx`	Vector of lambda indices on which the prediction is to be made. If not provided, will predict on all lambdas found.
`family`	Type of the phenotype: "gaussian" for continuous phenotype and "binomial" for binary phenotype.
`status_col`	Name of the status column for Cox model.
`snpnet_prefix`	Prefix of the snpnet result files used to construct the full path. Only if 'saved_path' is specified.
`snpnet_suffix`	Extension suffix of the snpnet result files used to construct the full path. Only if 'saved_path' is specified.
`snpnet_subdir`	Name of the snpnet result subdirectory holding multiple result files for one phenotype. Only if 'saved_path' is specified.
`configs`	Additional list of configs including path to either zstdcat or zcat.