Showing papers by "Gianluca Pollastri published in 2010"

Improving the analysis of NMR spectra tracking pH‐induced conformational changes: Removing artefacts of the electric field on the NMR chemical shift

Abstract: pH-induced chemical shift perturbations (CSPs) can be used to study pH-dependent conformational transitions in proteins. Recently, an elegant principal component analysis (PCA) algorithm was developed and used to study the pH-dependent structural transitions in bovine beta-lactoglobulin (betaLG) by analyzing its NMR pH-titration spectra. Here, we augment this analysis method by filtering out changes in the NMR chemical shift that stem from effects that are electrostatic in nature. Specifically, we examine how many CSPs can be explained by purely electrostatic effects arising from titrational events in betaLG. The results show that around 20% of the amide nuclei CSPs in betaLG originate exclusively from "through-space" electric field effects. A PCA of NMR data where electric field artefacts have been removed gives a different picture of the pH-dependent structural transitions in betaLG. The method implemented here is well suited to be applied on a whole range of proteins, which experience at least one pH-dependent conformational change. Proteins 2010. (c) 2009 Wiley-Liss, Inc.

De Novo protein subcellular localization prediction by N-to-1 neural networks

Abstract: Knowledge of the subcellular location of a protein provides valuable information about its function and possible interaction with other proteins. In the post-genomic era, fast and accurate predictors of subcellular location are required if this abundance of sequence data is to be fully exploited. We have developed a subcellular localization predictor (SCL pred) which predicts the location of a protein into four classes for animals and fungi and five classes for plants (secretory pathway, cytoplasm, nucleus, mitochondrion and chloroplast) using high throughput machine learning techniques trained on large non-redundant sets of protein sequences. The algorithm powering SCL pred is a novel Neural Network (N-to-1 Neural Network, or N1-NN) which is capable of mapping whole sequences into single properties (a functional class, in this work) without resorting to predefined transformations, but rather by adaptively compressing the sequence into a hidden feature vector. We benchmark SCL pred against other publicly available predictors using two benchmarks including a new subset of Swiss-Prot release 57. We show that SCL pred compares favourably to the other state-of-the-art predictors. Moreover, the N1-NN algorithm is fully general and may be applied to a host of problems of similar shape, that is, in which a whole sequence needs to be mapped into a fixed-size array of properties, and the adaptive compression it operates may even shed light on the space of protein sequences. The predictive systems described in this paper are publicly available at http://distill.ucd.ie/distill/.