hirst2018: Data from Hirst et al. 2018

In lgatto/pRolocdata: Data accompanying the pRoloc package

Data from Hirst et al. 2018

Description

From the supplementary file notes:

These are the SILAC ratio data from 2046 proteins with complete profiles across all nine organellar maps.

Each profile consists of five ratios, corresponding to five subcellular fractions obtained by differential centrifugation (3000 x g pellet, 6000 x g pellet, 12000 x g pellet, 24000 x pellet, 80000 x g pellet). The centrifugation speeds are available centriguation in the MSnSet object.

Each ratio shows the abundance of the total membrane SILAC heavy spike-in relative to the abundance in a given subfraction.

Maps were made from three cell lines (control HeLa, and two independent AP5Z1 KO HeLa cell lines, called AP5KNC2 and AP5KOC6), each in triplicate (replicates R1, R2, and R3). The sample are code as "CTRL" (HeLa control), "C2" (AP5KNC2 AP5Z1 KO cells) and "C6" (AP5KNC6 AP5Z1 KO cells).

Marker proteins used to define organellar clusters in Supplemental Figure 1 in the manuscript are annotated as feature variable markers.

Finally, the ratios in the hirst2018 data where normalised by their sum (using normalise(, method = "sum")).

The feature data also contains information about the comparison of organellar maps made from control or AP5 ablated cells, revealing putative proteins that undergo subcellular localisation shifts. Each protein receives an M score (magnitude of movement), and an R score (reproducibility of movement, i.e. correlation between replicates). In addition, the reproducibility of movement between the two AP5 KO cell lines is scored (Correlation C2 vs C6). Note however that the authors themselves claim that:

‘The cutoffs chosen in Fig 1C (M > 1.5, R > 0.5) correspond to an estimated FDR of 23%. Please note that the actual FDR is probably lower than this estimated FDR, because the mock data lack the additional cell line and the clonal correlation filter.’

The re-localisation candidates are those that have an M score > 1.5 and a R score > 0.5, and are marked with a hit feature variable set to TRUE.

Usage

data(hirst2018)

Format

The data is an instance of class MSnSet from package MSnbase.

References

Hirst J, Itzhak DN, Antrobus R, Borner GHH, Robinson MS. Role of the AP-5 adaptor protein complex in late endosome-to-Golgi retrieval. PLoS Biol. 2018 Jan 30;16(1):e2004411. doi: 10.1371/journal.pbio.2004411. eCollection 2018 Jan. PubMed PMID: 29381698; PubMed Central PMCID: PMC5806898.

Examples

## load the two 24 hours datasets
data(hirst2018)
hirst2018

## experimental design
table(pData(hirst2018)[, -2])

## the expression data
exprs(hirst2018)[1:5, 1:3]

## abstract
abstract(hirst2018)

## split data by samples
x <- split(hirst2018, "sample")

## These are the relocalisation hits
hits <- which(fData(hirst2018)$Hits)
reloc <- FeaturesOfInterest(description = "Relocation hits",
			    featureNames(hirst2018)[hits])
reloc

## plotting
library("pRoloc")
par(mfrow = c(1, 3))
plot2D(x[[1]], main = "AP5KNC2")
highlightOnPlot(x[[1]], reloc)
plot2D(x[[2]], main = "AP5KNC6")
highlightOnPlot(x[[1]], reloc)
plot2D(x[[3]], main = "HeLa control")
highlightOnPlot(x[[1]], reloc)
addLegend(x[[3]], where = "topleft")

lgatto/pRolocdata documentation built on Oct. 17, 2024, 10:16 p.m.