A novel ground truth multispectral image dataset with weight, anthocyanins, and Brix index measures of grape berries tested for its utility in machine learning pipelines.

BACKGROUND: The combination of computer vision devices such as multispectral cameras coupled with artificial intelligence has provided a major leap forward in image-based analysis of biological processes. Supervised artificial intelligence algorithms require large ground truth image datasets for model training, which allows to validate or refute research hypotheses and to carry out comparisons between models. However, public datasets of images are scarce and ground truth images are surprisingly few considering the numbers required for training algorithms. RESULTS: We created a dataset of 1,283 multidimensional arrays, using berries from five different grape varieties. Each array has 37 images of wavelengths between 488.38 and 952.76 nm obtained from single berries. Coupled to each multispectral image, we added a dataset with measurements including, weight, anthocyanin content, and Brix index for each independent grape. Thus, the images have paired measures, creating a ground truth dataset. We tested the dataset with 2 neural network algorithms: multilayer perceptron (MLP) and 3-dimensional convolutional neural network (3D-CNN). A perfect (100% accuracy) classification model was fit with either the MLP or 3D-CNN algorithms. CONCLUSIONS: This is the first public dataset of grape ground truth multispectral images. Associated with each multispectral image, there are measures of the weight, anthocyanins, and Brix index. The dataset should be useful to develop deep learning algorithms for classification, dimensionality reduction, regression, and prediction analysis.

PhageCocktail: An R package to design phage cocktails from experimental phage-bacteria infection networks.

BACKGROUND AND OBJECTIVE: Phage therapy is a resurgent strategy used in medicine and the food industry to lyse bacteria that cause damage to health or spoil a food product. Frequently, phage-bacteria infection networks have a large size, making it impossible to manually study all possible phage cocktails. Thus, this article presents an R package called PhageCocktail to automatically design efficient phage cocktails from phage-bacteria infection networks. METHODS: This R package includes four different methods for designing phage cocktails: ExhaustiveSearch, ExhaustivePhi, ClusteringSearch, and ClusteringPhi. These four methods are explained in detail and are evaluated using 13 empirical phage-bacteria infection networks. More specifically, runtime and expected success (fraction of lysed bacteria) are analyzed. RESULTS: The four methods have variations in terms of runtime and quality of the results. ExhaustiveSearch always provides the best possible phage cocktail, but its runtime could be long. ExhaustivePhi only focuses on one cocktail size, the one estimated as the best; thus, its runtime is less than ExhaustiveSearch, but it can produce cocktails with more phages than necessary. ClusteringSearch and ClusteringPhi are very fast (generally, less than one millisecond), providing always immediate results due to clustering techniques, but their accuracies can be lower, yielding cocktails with lower expected successes. CONCLUSIONS: The larger the phage-bacteria infection network is, the more complex its analysis is. Thus, this tool eases this task for scientists and other users while designing phage cocktails of good quality. This R package includes four different methods; therefore, users may choose among them, considering their preferences in speed and accuracy of results.