Abstract
The discovery of new biological interactions, such as interactions between drugs and cell lines, can improve the way drugs are developed. Recently, there has been important interest for predicting interactions between drugs and targets using recommender systems; and more specifically, using recommender systems to predict drug activity on cellular lines. In this work, we present a simple and straightforward approach for the discovery of interactions between drugs and cellular lines using collaborative filtering. We represent cellular lines by their drug affinity profile, and correspondingly, represent drugs by their cell line affinity profile in a single interaction matrix. Using simple matrix factorization, we predicted previously unknown values, minimizing the regularized squared error. We build a comprehensive dataset with information from the ChEMBL database. Our dataset comprises 300,000+ molecules, 1,200+ cellular lines, and 3,000,000+ reported activities. We have been able to successfully predict drug activity, and evaluate the performance of our model via utility, achieving an Area Under ROC Curve (AUROC) of near 0.9. Copyright

Abstract
We implement a quantum binary classifier where given a dataset of pairs of training inputs and target outputs our goal is to predict the output of a new input. The script is based in a hybrid scheme inspired in an existing PennyLane's variational classifier and to encode the classical data we resort to PennyLane's amplitude encoding embedding template. We use the quantum binary classifier applied to the well known Iris dataset and to a car traffic dataset. Our results show that the quantum approach is capable of performing the task using as few as 2 qubits. Accuracies are similar to other quantum machine learning research studies, and as good as the ones produced by classical classifiers.

Abstract
This work discusses a strategy named Map, Optimize and Learn (MOL) which analyzes how to change the representation of samples of a 2D dataset to generate useful patterns for classification tasks using Convolutional Neural Networks (CNN) architectures. The strategy is applied to a real-world scenario of children and teenagers with cardiac pathology and compared against state of the art Machine Learning (ML) algorithms for 2D datasets. Preliminary results suggests that the strategy has potential to improve the prediction quality.

Abstract
This work presents the mapping of the traveling salesperson problem (TSP) based in pseudo-Boolean constraints to a graph of the D-Wave Systems Inc. We first formulate the problem as a set of constraints represented in propositional logic and then resort to the SATyrus approach to convert the set of constraints to an energy minimization problem. Next, we transform the formulation to a quadratic unconstrained binary optimization problem (QUBO) and solve the problem using different approaches: (a) classical QUBO using simulated annealing in a von Neumann machine, (b) QUBO in a simulated quantum environment, (c) QUBO using the D-Wave quantum machine. Moreover, we study the amount of time and execution time reduction we can achieve by exploring approximate solutions using the three approaches. Results show that for every graph size tested with the number of nodes less than or equal to 7, we can always obtain at least one optimal solution. In addition, the D-Wave machine can find optimal solutions more often than its classical counterpart for the same number of iterations and number of repetitions. Execution times, however, can be some orders of magnitude higher than the classical or simulated approaches for small graphs. For a higher number of nodes, the average execution time to find the first optimal solution in the quantum machine is 26% (n = 6) and 47% (n = 7) better than the classical.

Abstract
Association football has been the subject of many research studies. In this work we present a study on player similarity using passing sequences extracted from games from the top-5 European football leagues during the 2017/2018 season. We present two different approaches: first, we only count the motifs a player is involved in; then we also take into consideration the specific position a player occupies in each motif. We also present a new way to objectively judge the quality of the generated models in football analytics. Our results show that the study of passing sequences can be used to study player similarity with relative success. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Abstract