Arguments For A TWAS Weight-Fitting Engine

Options forwarded to the engine behind a TWAS weight method, checked against that engine's live formals where it enumerates them. Pass the result as an element of TwasWeightsMethodsParam, or directly to the matching weights function.

A method's individual-level and summary-statistic paths often run DIFFERENT engines, and then each takes its own constructor: lassoWeights fits with glmnet (GlmnetOptions) while lassosumRssWeights runs pecotmr's lassosum solver (LassosumOptions); scadWeights / mcpWeights / l0learnWeights use ncvreg and L0Learn while their *Rss counterparts use PenalizedRssOptions; dprGibbsWeights uses RcppDPR and sdprWeights uses SdprOptions. Splitting them is what lets each be checked exactly, rather than against a union in which an option meant for the other path would pass and then be dropped.

GlmnetOptions, NcvregOptions and DprOptions cannot check for an UNKNOWN name: cv.glmnet, cv.ncvreg and fit_model each take ..., so they accept any name and there is nothing to check against. Every other constructor here rejects an unknown option. Separately, and in all of them, a name pecotmr supplies itself is refused — GlmnetOptions(alpha=) and NcvregOptions(penalty=) select the penalty family, which is the calling wrapper's to set. See MethodOptions.

GlmnetOptions rejects two names by hand, which the ... check above would otherwise let through in silence. nfold is not a cv.glmnet argument at all — the fold count is nfolds, and the shorter spelling only ever worked through R's partial matching. parallel is a real one, but pecotmr does not expose it: glmnet parallelises over its CV folds through a foreach backend that this package neither declares nor registers, and weight fitting already runs inside BiocParallel workers over the same folds, so enabling it would nest two frameworks and oversubscribe the machine. The glasso and univariate-glmnet mr.mash initialisations are single-threaded for the same reason.

Usage

GlmnetOptions(...)

QggOptions(...)

BglrOptions(...)

DprOptions(...)

NcvregOptions(...)

L0learnOptions(...)

MrashOptions(...)

MrashRssOptions(...)

PrsCsOptions(...)

LassosumOptions(...)

PenalizedRssOptions(...)

SdprOptions(...)

MrmashOptions(...)

MrmashRssOptions(...)

Arguments

...

Arguments for the engine, under its own names.

Value

A MethodOptions object.

Examples

GlmnetOptions(nfolds = 10)
#> <MethodOptions: glmnet>
#>   forwarded to glmnet::cv.glmnet()
#>   argument names NOT checked: glmnet::cv.glmnet() takes `...`, so any name is accepted
#>   for input class QtlDataset
#>   nfolds                 10
QggOptions(updateB = TRUE)
#> <MethodOptions: qgg>
#>   forwarded to qgg::gbayes()
#>   argument names checked against qgg::gbayes()
#>   for input class QtlDataset
#>   updateB                TRUE
LassosumOptions(thr = 1e-4)
#> <MethodOptions: lassosum>
#>   forwarded to pecotmr::lassosumRss()
#>   argument names checked against pecotmr::lassosumRss()
#>   for input class QtlSumStats
#>   thr                    1e-04