SUMMARY: The ability to unveil binding patterns in peptide sets has important applications in several biomedical areas, including the development of vaccines. We present an open-source tool, CNN-PepPred, that uses convolutional neural networks to discover such patterns, along with its application to peptide-HLA class II binding prediction. The tool can be used locally on different operating systems, with CPUs or GPUs, to train, evaluate, apply and visualize models. AVAILABILITY AND IMPLEMENTATION: CNN-PepPred is freely available as a Python tool with a detailed User's Guide at https://github.com/ComputBiol-IBB/CNN-PepPred. The site includes the peptide sets used in this study, extracted from IEDB (www.iedb.org). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Cross-(multi)platform normalization of gene-expression microarray data remains an unresolved issue. Despite the existence of several algorithms, they are either constrained by the need to normalize all samples of all platforms together, compromising scalability and reuse, by adherence to the platforms of a specific provider, or simply by poor performance. In addition, many of the methods presented in the literature have not been specifically tested against multi-platform data and/or other methods applicable in this context. Thus, we set out to develop a normalization algorithm appropriate for gene-expression studies based on multiple, potentially large microarray sets collected along multiple platforms and at different times, applicable in systematic studies aimed at extracting knowledge from the wealth of microarray data available in public repositories; for example, for the extraction of Real-World Data to complement data from Randomized Controlled Trials. Our main focus or criterion for performance was on the capacity of the algorithm to properly separate samples from different biological groups. RESULTS: We present CuBlock, an algorithm addressing this objective, together with a strategy to validate cross-platform normalization methods. To validate the algorithm and benchmark it against existing methods, we used two distinct datasets, one specifically generated for testing and standardization purposes and one from an actual experimental study. Using these datasets, we benchmarked CuBlock against ComBat (Johnson et al., 2007), UPC (Piccolo et al., 2013), YuGene (Lê Cao et al., 2014), DBNorm (Meng et al., 2017), Shambhala (Borisov et al., 2019) and a simple log2 transform as reference. We note that many other popular normalization methods are not applicable in this context. CuBlock was the only algorithm in this group that could always and clearly differentiate the underlying biological groups after mixing the data, from up to six different platforms in this study. AVAILABILITY AND IMPLEMENTATION: CuBlock can be downloaded from https://www.mathworks.com/matlabcentral/fileexchange/77882-cublock. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Abstract Summary: The regulation of protein activity is a key aspect of life at the molecular level. Unveiling its details is thus crucial to understanding signalling and metabolic pathways. The most common and powerful mechanism of protein-function regulation is allostery, which has been increasingly calling the attention of medicinal chemists due to its potential for the discovery of novel therapeutics. In this context, PARS is a simple and fast method that queries protein dynamics and structural conservation to identify pockets on a protein structure that may exert a regulatory effect on the binding of a small-molecule ligand. Availability: PARS is freely available as a web server at http://bioinf.uab.cat/pars. Contact: [email protected] Supplementary information: Supplementary data are available at Bioinformatics online.

The pilus 2a backbone protein (BP-2a) is one of the most structurally and functionally characterized components of a potential vaccine formulation against Group B Streptococcus. It is characterized by six main immunologically distinct allelic variants, each inducing variant-specific protection. To investigate the molecular determinants driving the variant immunogenic specificity of BP-2a, in terms of single residue contributions, we generated six monoclonal antibodies against a specific protein variant based on their capability to recognize the polymerized pili structure on the bacterial surface. Three mAbs were also able to induce complement-dependent opsonophagocytosis killing of live GBS and target the same linear epitope present in the structurally defined and immunodominant domain D3 of the protein. Molecular docking between the modelled scFv antibody sequences and the BP-2a crystal structure revealed the potential role at the binding interface of some non-conserved antigen residues. Mutagenesis analysis confirmed the necessity of a perfect balance between charges, size and polarity at the binding interface to obtain specific binding of mAbs to the protein antigen for a neutralizing response.

BACKGROUND: Allostery is one of the most powerful and common ways of regulation of protein activity. However, for most allosteric proteins identified to date the mechanistic details of allosteric modulation are not yet well understood. Uncovering common mechanistic patterns underlying allostery would allow not only a better academic understanding of the phenomena, but it would also streamline the design of novel therapeutic solutions. This relatively unexplored therapeutic potential and the putative advantages of allosteric drugs over classical active-site inhibitors fuel the attention allosteric-drug research is receiving at present. A first step to harness the regulatory potential and versatility of allosteric sites, in the context of drug-discovery and design, would be to detect or predict their presence and location. In this article, we describe a simple computational approach, based on the effect allosteric ligands exert on protein flexibility upon binding, to predict the existence and position of allosteric sites on a given protein structure. RESULTS: By querying the literature and a recently available database of allosteric sites, we gathered 213 allosteric proteins with structural information that we further filtered into a non-redundant set of 91 proteins. We performed normal-mode analysis and observed significant changes in protein flexibility upon allosteric-ligand binding in 70% of the cases. These results agree with the current view that allosteric mechanisms are in many cases governed by changes in protein dynamics caused by ligand binding. Furthermore, we implemented an approach that achieves 65% positive predictive value in identifying allosteric sites within the set of predicted cavities of a protein (stricter parameters set, 0.22 sensitivity), by combining the current analysis on dynamics with previous results on structural conservation of allosteric sites. We also analyzed four biological examples in detail, revealing that this simple coarse-grained methodology is able to capture the effects triggered by allosteric ligands already described in the literature. CONCLUSIONS: We introduce a simple computational approach to predict the presence and position of allosteric sites in a protein based on the analysis of changes in protein normal modes upon the binding of a coarse-grained ligand at predicted cavities. Its performance has been demonstrated using a newly curated non-redundant set of 91 proteins with reported allosteric properties. The software developed in this work is available upon request from the authors.

BACKGROUND: Protein aggregation correlates with the development of several debilitating human disorders of growing incidence, such as Alzheimer's and Parkinson's diseases. On the biotechnological side, protein production is often hampered by the accumulation of recombinant proteins into aggregates. Thus, the development of methods to anticipate the aggregation properties of polypeptides is receiving increasing attention. AGGRESCAN is a web-based software for the prediction of aggregation-prone segments in protein sequences, the analysis of the effect of mutations on protein aggregation propensities and the comparison of the aggregation properties of different proteins or protein sets. RESULTS: AGGRESCAN is based on an aggregation-propensity scale for natural amino acids derived from in vivo experiments and on the assumption that short and specific sequence stretches modulate protein aggregation. The algorithm is shown to identify a series of protein fragments involved in the aggregation of disease-related proteins and to predict the effect of genetic mutations on their deposition propensities. It also provides new insights into the differential aggregation properties displayed by globular proteins, natively unfolded polypeptides, amyloidogenic proteins and proteins found in bacterial inclusion bodies. CONCLUSION: By identifying aggregation-prone segments in proteins, AGGRESCAN http://bioinf.uab.es/aggrescan/ shall facilitate (i) the identification of possible therapeutic targets for anti-depositional strategies in conformational diseases and (ii) the anticipation of aggregation phenomena during storage or recombinant production of bioactive polypeptides or polypeptide sets.