BastionHub: a universal platform for integrating and analyzing substrates secreted by Gram-negative bacteria

Abstract

Gram-negative bacteria utilize secretion systems to export substrates into their surrounding environment or directly into neighboring cells. These substrates are proteins that function to promote bacterial survival: by facilitating nutrient collection, disabling competitor species or, for pathogens, to disable host defenses. Following a rapid development of computational techniques, a growing number of substrates have been discovered and subsequently validated by wet lab experiments. To date, several online databases have been developed to catalogue these substrates but they have limited user options for in-depth analysis, and typically focus on a single type of secreted substrate. We therefore developed a universal platform, BastionHub, that incorporates extensive functional modules to facilitate substrate analysis and integrates the five major Gram-negative secreted substrate types (i.e. from types I–IV and VI secretion systems). To our knowledge, BastionHub is not only the most comprehensive online database available, it is also the first to incorporate substrates secreted by type I or type II secretion systems. By providing the most up-to-date details of secreted substrates and state-of-the-art prediction and visualized relationship analysis tools, BastionHub will be an important platform that can assist biologists in uncovering novel substrates and formulating new hypotheses. BastionHub is freely available at http://bastionhub.erc.monash.edu/.

α-Helices in the Type III Secretion Effectors: A Prevalent Feature with Versatile Roles

“…Representative T3SEs were selected through the BastionHub catalogue [ 116 ] for type III secreted effectors with Protein Data Bank (PDB) records and unique folds. Data mining from the above databases took place on January 2021.…”

DeepT3 2.0: improving type III secreted effector predictions by an integrative deep learning framework

“…More recently, Wang et al. proposed a universal platform of Gram-negative secreted substrates-BastionHub, which is not only the most comprehensive online database but also helpful for in-depth analysis of five major types of secreted substrates such as the type I, II, III, IV and VI secreted proteins ( 60 ). Furthermore, there have been several excellent reviews on machine learning algorithms and protein characterization methods for T3SE predictions ( 61–64 ).…”

PncsHub: a platform for annotating and analyzing non-classically secreted proteins in Gram-positive bacteria

“…PncsHub is based largely on what was previously implemented for AcrHub and BastionHub ( 25 , 59 ) in terms of website design and implementation, unless otherwise indicated. It was implemented as three separate modules: an Investigation Module, a Prediction Module, and a Relationship Analysis Module (Figure 1 ).…”

“…Of the two most recent webservers, PeNGaRoo and NonClasGP-Pred, both have opted for a binary output (yes/no) and are therefore not capable of determining which non-classical secretion system the protein is likely secreted by. Here, we describe the Gram- p ositive n on- c lassically s ecreted (Pncs) protein Hub : PncsHub (we pronounce it ‘Pinks Hub’) (Figure 1 ), a companion database to BastionHub ( 25 ), which catalogues Gram-negative non-classically secreted proteins instead. PncsHub combines high quality prediction algorithms, catalogues experimentally verified proteins, and provides a series of data analysis and visualisation tools that can all be used to facilitate both discovery and annotation, and ultimately allow users to determine which secretion pathway a putative substrate likely uses.…”