baers2018: Synechocystis spatial proteomics

In lgatto/pRolocdata: Data accompanying the pRoloc package

Synechocystis spatial proteomics

Description

Data from 'Spatial mapping of a cyanobacterial proteome reveals distinct subcellular compartment organisation and dynamic metabolic pathways' (submitted).

Cyanobacteria are complex prokaryotes, incorporating a Gram-negative cell wall and internal thylakoid membranes. However, localisation of proteins within cyanobacterial cells is poorly understood. Using subcellular fractionation and quantitative proteomics we report the most extensive subcellular map of the proteome of a cyanobacterial cell, identifying ~67% of Synechocystis sp. PCC 6803 proteins, ~1000 more than previous studies. 1711 proteins were assigned to six specific subcellular regions.

This dataset is composed of two combined replicated 10-plex LOPIT experiments.

Protein markers for the plasma membrane, thylakoid membrane, cytosol, and small and la rge ribosomal subunits were curated from a literature review. A Support Vector Machine (SVM) classifier was employed on the combined dataset, with a radial basis function kernel, using class specific weights for classification of unassigned proteins to one of the 5 defined sub-cellular niches, TM, PM, soluble, small ribosomal subunit, large ribosomal subunit. The weights used in classification were set to be inversely proportional to the subcellular class frequencies to account for class imbalance. Algorithmic performance of the SVM on the dataset was estimated (as described in Trotter et al 2010). Scoring thresholds were calculated per subcellular niche and were set based on concordance with existing subcellular knowledge annotation to attain a 7.5% false discovery rate (FDR). Unassigned proteins were then classified to 1 of the 5 compartments according to the SVM prediction if greater than the calcul ated class threshold.

Usage

data("baers2018")

Examples

data(baers2018)

library("pRoloc")
par(mfrow = c(1, 2))
plot2D(baers2018, main = "Markers")
addLegend(baers2018, where = "bottomright")
plot2D(baers2018, fcol = "Localisation", main = "Localisation")

lgatto/pRolocdata documentation built on April 7, 2023, 1:56 a.m.