2

Here we generalize property-based encoders for amino acid sequences and demonstrate that they substantially increase ML models’ performance in predictive tasks in Protein Engineering. These encoders arise from semantically-consistent groups of properties within the AAIndex database and, combined with Fourier Transforms, increase the performance metrics across various tasks. All these results offer simple ways to increase the performance of general and complex predictive tasks in protein engineering without increasing their complexity.

At the beginning of the COVID-19 pandemic, very little was known about both the disease and the virus that caused it. As more information became available, the public awareness reached unprecedented scales; epidemiological terms such as incidence or reproduction number infected our daily conversations. Consequently, high pressure fell on the shoulders of policymakers, who were expected to point us to the way out of this global health threat. However, whom do we ask when we all are still learning? We would say, “let us build a model!”.

The emergence of SARS-CoV-2 variants with enhanced transmissibility and partial immune escape (also known as Variants of Concern, VOC) has characterised the second year of the COVID-19 pandemic. Genomic surveillance allows identifying hidden, fast-spreading lineages and VOCs before they become a public health threat.

We use a mathematical model to represent individual OGTT curves as the coordinated action of subsystems, each one described by a parameter with physiological interpretation. This method allows identifying individuals at high risk of dysglycemia—which would have been missed with traditional binary diagnostic methods—enabling early nutritional intervention with a polyphenol-supplemented diet where it is most effective and desirable.

Pharmaceutical and non-pharmaceutical interventions (NPIs) have been crucial for controlling COVID-19. They are complemented by voluntary health-protective behavior, building a complex interplay between risk perception, behavior, and disease spread. We studied how voluntary health-protective behavior and vaccination willingness impact the long-term dynamics. Altogether, our work highlights the importance of appropriate mandatory NPIs to maximise the impact of individual voluntary actions in pandemic control.

The COVID-19 pandemic changed nearly every aspect of our lives. The rapid spread of the disease exposed several layers of inequality that it has exploited to propagate preferentially. We find these layers in different contexts and levels, such as the impossibility to self-isolate and do remote work (at the individual level) or economic constraints to deploy a fast vaccination program (at the country level). In the context of vaccination programmes, resources must be optimised to alleviate the pandemic burden where it is needed the most.

The traditional long-term solutions for epidemic control involve eradication or population immunity. Here, we analytically derive the existence of a third viable solution; a stable equilibrium at low case numbers, where test-trace-and-isolate policies partially compensate for local spreading events and only moderate restrictions remain necessary.

false

false

false