In Bioconductor/DelayedDataFrame: Delayed operation on DataFrame using standard DataFrame metaphor

DelayedDataFrame class

• Extends: DataFrame

library(DelayedDataFrame)
(ddf <- DelayedDataFrame())

• lazyIndex slot
  Saves the mapping indexes for each column.

lazyIndex(ddf)

Constructor:

Each argument in "..." is coerced to a DataFrame and combined column-wise.

DelayedDataFrame(..., row.names=FALSE, check.names=TRUE)

(obj <- DelayedDataFrame(letters, LETTERS, row.names=LETTERS))

---

lazyIndex(obj)

On-disk data representation in DataFrame format

• Genomic Data Structure (GDS)

CoreArray Genomic Data Structure (GDS) is designed for large-scale datasets (for available random-access memory). The Bioconductor package gdsfmt has provided a high-level R interface to GDS.

---

file <- SeqArray::seqExampleFileName("gds")
f <- gdsfmt::openfn.gds(file)
f

closefn.gds(f)

---

• GDSArray

GDSArray is an R and Bioconductor package, that represents GDS files as DelayedArray instances.

library(GDSArray)
gdsnodes(file)

---

Use GDSArray to represent the GDS nodes for variant annotation.

varid <- GDSArray(file, "annotation/id")
AA <- GDSArray(file, "annotation/info/AA")

varid

---

seed(varid)

---

Construct a DelayedDataFrame object with GDSArray columns.

(ddf <- DelayedDataFrame(varid, AA))

subsetting

• listData: unchanged

(ddf1 <- ddf[1:20,])
identical(ddf@listData, ddf1@listData)

---

• lazyIndex: updated
• nrows: updated
• rownames: updated (if not NULL)

lazyIndex(ddf1)
nrow(ddf1)

lazyIndex realization

as(ddf1, "DataFrame")

Availability

The development version is available to download through github:

devtools::install_github("Bioconductor/DelayedDataFrame")

Bioconductor/DelayedDataFrame documentation built on Oct. 29, 2023, 5:02 p.m.