RunLassosum2Rule

lassosum2 model rule

Description

RunLassosum2Rule resolves one polygenius.model output for lassosum2.

Details

Reactive rule for model outputs generated by algorithm lassosum2.

The rule requests GWAS input at required pval.max, binds an ld.bigsnpr storage resource for the matched panel/build, matches GWAS variants to that storage in-process, runs lassosum2, and returns one deterministic model.

User-facing LD construction controls map onto ld.bigsnpr storage fields:

• ld.size -> window

• ld.thr -> threshold

These controls affect how the upstream bigsnpr LD storage is built. They are separate from lassosum2 fitting hyperparameters such as delta and lambda.

The lassosum2 workflow fixes bigsnpr LD fields statistic = “r”, signed = TRUE, and phased = FALSE.

The bigsnpr tutorial treats the returned lassosum2 matrix as a grid of candidate models and chooses among columns using prediction performance in a validation set. This rule does not receive validation genotypes or outcomes, so validation-set choice is left to downstream evaluation workflows.

Super class

PolyGenius::Rule -> RunLassosum2Rule

Methods

Public methods

• RunLassosum2Rule$new()

• RunLassosum2Rule$matches()

• RunLassosum2Rule$inputs()

• RunLassosum2Rule$bind()

• RunLassosum2Rule$requirements()

• RunLassosum2Rule$run()

• RunLassosum2Rule$clone()

Inherited methods

• PolyGenius::Rule$print()

---

Method new()

Create the lassosum2 rule.

Usage

RunLassosum2Rule$new()

---

Method matches()

Check whether this rule can produce the requested output.

Usage

RunLassosum2Rule$matches(output.spec)

Arguments

output.spec

Output resource specification.

Returns

TRUE for lassosum2 model outputs.

---

Method inputs()

Build input specifications for this output.

Usage

RunLassosum2Rule$inputs(output.spec)

Arguments

output.spec

Output resource specification.

Returns

Named list with required GWAS input and deferred LD input.

---

Method bind()

Bind LD storage input after GWAS build is known.

Usage

RunLassosum2Rule$bind(inputs, store, output.spec = NULL)

Arguments

inputs

Named list of input specifications.

store

Resource store.

output.spec

Output resource specification.

Returns

Updated named list of inputs.

---

Method requirements()

Declare resource requirements for this rule.

Usage

RunLassosum2Rule$requirements(output.spec, inputs)

Arguments

output.spec

Output resource specification.

inputs

Bound input values/specifications.

Returns

Named list with cores and memory.

---

Method run()

Execute lassosum2 model construction.

Usage

RunLassosum2Rule$run(output.spec, inputs, logger)

Arguments

output.spec

Output resource specification.

inputs

Named list of bound input values.

logger

Rule logger.

Returns

Rule result list.

---

Method clone()

The objects of this class are cloneable with this method.

Usage

RunLassosum2Rule$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.