Accelerating Computational Materials Discovery with Machine Learning and Cloud High-Performance Computing: from Large-Scale Screening to Experimental Validation.

High-throughput computational materials discovery has promised significant acceleration of the design and discovery of new materials for many years. Despite a surge in interest and activity, the constraints imposed by large-scale computational resources present a significant bottleneck. Furthermore, examples of very large-scale computational discovery carried out through experimental validation remain scarce, especially for materials with product applicability.

Impossibility theorems for feature attribution.

Despite a sea of interpretability methods that can produce plausible explanations, the field has also empirically seen many failure cases of such methods. In light of these results, it remains unclear for practitioners how to use these methods and choose between them in a principled way. In this paper, we show that for moderately rich model classes (easily satisfied by neural networks), any feature attribution method that is complete and linear-for example, Integrated Gradients and Shapley Additive Explanations (SHAP)-can provably fail to improve on random guessing for inferring model behavior.

Inadequate Measles, Mumps, Rubella, and Varicella Immunity Among Employees.

Objectives: The purpose of this cohort study was to evaluate measles, mumps, rubella (MMR), and varicella immunity among a population of adult employees receiving primary care in an employer-sponsored health center.

Methods: Participants were eligible for MMR and varicella immunity screening if they were an employee receiving primary care in an employer-sponsored health center between January 1, 2019 and November 1, 2020 who could not provide proof of immunization and 1) had it recommended by their provider, 2) specifically requested immunity testing (often because they had heard of measles outbreaks in their country of origin), or 3) were seen for an immigration physical for their Green Card application.

Results: Overall, 3494 patients were screened for their MMR immunity.

Cognitive evolution in natalizumab-treated multiple sclerosis patients.

Background: Cognitive dysfunction affects up to 65% of multiple sclerosis (MS) patients and progresses over time. Natalizumab has been shown to be superior to placebo in preserving cognition for the first two years of therapy.

Objectives: The objectives of this study are to understand the impact of natalizumab on cognition beyond two years of therapy and to investigate whether baseline characteristics are predictive of clinical response.

[French-Canadian validation of the MOS Social Support Survey].

Background: The MOS Social Support Survey was developed for patients who participated in the Medical Outcomes Study (MOS), a two-year study on persons who suffered from chronic illness. There are a number of advantages to using the MOS Social Support Survey, especially with those persons who suffer from chronic illness: it is easy to understand, is relatively short, is multidimensional, can be completed by the patient without assistance and has good psychometric properties.

Objective: The goal of this study was to establish a French-Canadian version of the MOS Social Support Survey and to verify its psychometric properties following the cross-cultural translation and validation procedures proposed by Vallerand.