MSOmicsDesign

Note: This is a personal side project built for in-house experiment design.

MSOmicsDesign provides tools for designing mass spectrometry-based omics experiments. It covers two main workflows:

• Plate design — insert QC runs, block-randomise by experimental covariates, distribute samples across multi-well plates, visualise layouts, and export position lists for Thermo Xcalibur sequence files.
• DIA window design — generate fixed-width, variable-width (density-weighted), and gas-phase fractionation (GPF) isolation window lists with optional edge optimisation for reduced chimeric spectra.

---

Installation

# Using remotes
remotes::install_github("LKittinun/MSOmicsDesign")

# Using devtools
devtools::install_github("LKittinun/MSOmicsDesign")

---

Plate Design Workflow

library(MSOmicsDesign)

# 1. Define samples and insert QC every 6 injections
samples    <- paste0("S", 1:20)
samples_qc <- addQC(samples, interval = 6)

# 2. Fill into plates (6 × 9 by default, auto-spills to next plate)
fp <- fill.plate(samples_qc,
                 labels = paste0("lbl_", seq_along(samples_qc)))

# 3. Visualise — colour by group, show labels
plot(fp, labels = fp$label, control.suffix = "QC")

# 4. Export plate positions for Xcalibur
positions <- generate.position(fp, flatten = TRUE)

# 5. Write Xcalibur sequence file
write.sld(fp, filename = "sequence.csv")

Block randomisation

When your experiment has multiple covariates, block.rand() ensures balanced allocation across run-order blocks before filling plates.

data(ccc)
ccc_block <- block.rand(ccc, time, response)
check.block(ccc_block)           # inspect balance per block

fp_rand <- fill.plate(ccc_block$unique_group_id)
plot(fp_rand)

---

DIA Window Design Workflow

Fixed-width windows

# 30 equal windows between 400 – 1000 m/z, edges in forbidden zones
dia <- DIA.windows(mnc = 400, mxc = 1000, n.windows = 30)
plot(dia)

Variable-width windows (density-weighted)

Windows are narrower in dense precursor regions (uses built-in HeLa reference).

vdia <- vDIA.windows(mnc = 400, mxc = 1000, n.windows = 32)
plot(vdia)

Staggered windows

dia_stagger <- DIA.windows(mnc = 400, mxc = 1000, windows.width = 20,
                            stagger = TRUE)

GPF-DIA partitioning

Split an existing isolation list into per-injection GPF groups.

data(staggered_mz)
gpf <- GPF.windows(staggered_mz$range, mnc = 400, mxc = 1000)
names(gpf)      # one data frame per GPF group
plot(gpf[[1]])

---

Functions

Family	Function	Description
Plate	addQC()	Insert QC runs at intervals or positions
Plate	replaceQC()	Replace slots with QC labels (fixed length)
Plate	fill.plate()	Distribute samples across plate matrices
Plate	plot.plate_list()	Visualise a plate list
Plate	plot.plate()	Visualise a single plate
Plate	generate.position()	Export well positions for sequence files
Plate	write.sld()	Write Xcalibur .sld sequence CSV
Randomise	block.rand()	Block randomisation by covariates
Randomise	check.block()	Inspect covariate balance per block
m/z	DIA.windows()	Fixed-width DIA isolation windows
m/z	vDIA.windows()	Variable-width DIA isolation windows
m/z	GPF.windows()	GPF-DIA window partitioning
m/z	opt.windows()	Optimise window edge placement
m/z	plot.mzdf()	Visualise isolation windows

---

Vignettes

vignette("plate-design",  package = "MSOmicsDesign")
vignette("dia-windows",   package = "MSOmicsDesign")

---

License

CC BY 4.0 — Kittinun Leetanaporn