BionoiNet: ligand-binding site classification with off-the-shelf deep neural network.

MOTIVATION: Fast and accurate classification of ligand-binding sites in proteins with respect to the class of binding molecules is invaluable not only to the automatic functional annotation of large datasets of protein structures but also to projects in protein evolution, protein engineering and drug development. Deep learning techniques, which have already been successfully applied to address challenging problems across various fields, are inherently suitable to classify ligand-binding pockets. Our goal is to demonstrate that off-the-shelf deep learning models can be employed with minimum development effort to recognize nucleotide- and heme-binding sites with a comparable accuracy to highly specialized, voxel-based methods. RESULTS: We developed BionoiNet, a new deep learning-based framework implementing a popular ResNet model for image classification. BionoiNet first transforms the molecular structures of ligand-binding sites to 2D Voronoi diagrams, which are then used as the input to a pretrained convolutional neural network classifier. The ResNet model generalizes well to unseen data achieving the accuracy of 85.6% for nucleotide- and 91.3% for heme-binding pockets. BionoiNet also computes significance scores of pocket atoms, called BionoiScores, to provide meaningful insights into their interactions with ligand molecules. BionoiNet is a lightweight alternative to computationally expensive 3D architectures. AVAILABILITY AND IMPLEMENTATION: BionoiNet is implemented in Python with the source code freely available at: https://github.com/CSBG-LSU/BionoiNet. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Effect of complement component 5 polymorphisms on mastitis resistance in Egyptian buffalo and cattle.

Mastitis is one of the costliest diseases affecting the world's dairy industry. The important contribution of complement Component 5 (C5) to phagocytosis, which plays a major role in the defence of the bovine mammary gland against infection, makes this component of innate immunity a potential contributor in defending udder against mastitis. The objectives of this study were to sequence and analyse the whole coding region of the C5 gene in Egyptian buffalo and cattle, to detect any nucleotide variations (polymorphisms) and to investigate their associations with milk somatic cell score (SCS) as an indicator of mastitis in dairy animals. We sequenced a buffalo C5 cDNA fragment of 5336 bp (KP221293) and a cattle C5 cDNA fragment of 5303 bp (KP221294), which included the whole coding region and 3-UTR. Buffalo and cattle C5 cDNA shared sequence identity of 99%. The predicted complement C5 proteins consist of 1677 amino acid residues in both animals, one amino acid less than in humans and three amino acids more than in mouse C5 protein. Comparing cDNA sequences of different animals revealed nine novel SNPs in buffalo and seven SNPs in cattle, with two of them being novel. The association analysis revealed that five SNPs in buffalo are highly associated with SCS; indicating the contribution of complement C5 variants in buffalo mastitis resistance. No significant associations were detected between C5 variants and SCS in cattle. This is the first report about C5 variants in buffalo and its association with SCS.