RESUMO
Data sharing among different institutions represents one of the major challenges in developing distributed machine learning approaches, especially when data is sensitive, such as in medical applications. Federated learning is a possible solution, but requires fast communications and flawless security. Here, we propose SYNDSURV (SYNthetic Distributed SURVival), an alternative approach that simplifies the current state-of-the-art paradigm by allowing different centres to generate local simulated instances from real data and then gather them into a centralised hub, where an Artificial Intelligence (AI) model can learn in a standard way. The main advantage of this procedure is that it is model-agnostic, therefore prediction models can be directly applied in distributed applications without requiring particular adaptations as the current federated approaches do. To show the validity of our approach for medical applications, we tested it on a survival analysis task, offering a viable alternative to train AI models on distributed data. While federated learning has been mainly optimised for gradient-based approaches so far, our framework works with any predictive method, proving to be a comparable way of performing distributed learning without being too demanding towards each participating institute in terms of infrastructural requirements.
Assuntos
Inteligência Artificial , Aprendizado de Máquina , Análise de SobrevidaRESUMO
Missense variation in genomes can affect protein structure stability and, in turn, the cell physiology behavior. Predicting the impact of those variations is relevant, and the best-performing computational tools exploit the protein structure information. However, most of the current protein sequence variants are unresolved, and comparative or ab initio tools can provide a structure. Here, we evaluate the impact of model structures, compared to experimental structures, on the predictors of protein stability changes upon single-point mutations, where no significant changes are expected between the original and the mutated structures. We show that there are substantial differences among the computational tools. Methods that rely on coarse-grained representation are less sensitive to the underlying protein structures. In contrast, tools that exploit more detailed molecular representations are sensible to structures generated from comparative modeling, even on single-residue substitutions.
Assuntos
Biologia Computacional , Mutação Puntual , Biologia Computacional/métodos , Proteínas/metabolismo , Estabilidade Proteica , Sequência de AminoácidosRESUMO
The purpose of this study was to assess the myceliation rate, mycelial vigor and "estimated biomass" of Lentinula edodes (Berk.) Pegler, grown on a sugarcane bagasse substrate enriched with rice bran and sugarcane molasses for spawn production. The proportions of rice bran used were 0, 10, 15, 20, 25, 30 and 40 percent (dry weight/dry weight of bagasse) and the sugarcane molasses concentrations tested were 0, 10, 20, 30, 40, 50 and 60 g/kg (dry weight/dry weight of bagasse plus rice bran). The myceliation rate was decreased by the addition of the higher quantities of rice bran. The 25 and 30 percent rice bran proportions induced the highest stimulation of mycelial vigor. The addition of sugarcane molasses did not change myceliation rate or mycelial vigor. The "estimated biomass" values were similar when intermediate rice bran proportions were used and for all sugarcane molasses concentrations. Based on response surface obtained for the "estimated biomass" data, higher values were obtained with substrates containing 20 to 25 percent rice bran combined with 10 to 30 g sugarcane molasses, although the latter supplement was not considered to stimulate L. edodes growth.
Assuntos
Técnicas In Vitro , Melaço/análise , Micélio/crescimento & desenvolvimento , Micélio/isolamento & purificação , Cogumelos Shiitake , Saccharum/crescimento & desenvolvimento , Saccharum/microbiologia , Biomassa , MétodosRESUMO
This investigation was performed to evaluate the biological efficiency (BE), mean mushroom weight (MMW), mean number of mushroom (MNM) and mushroom quality of Shiitake [ Lentinula edodes (Berk.) Pegler] when grown on a sterilized substrate composed by sugarcane bagasse enriched with rice bran and sugarcane molasses. The proportions of rice bran were 0, 15, 20, 25 and 30 percent (dry weight/dry weight of bagasse); and the concentrations of sugarcane molasses were 0, 30 and 60 g/kg (dry weight/dry weight of bagasse plus rice bran). Four flushes were obtained during the production cycle, providing 3 accumulated productions which were used for production analysis. The substrate supplemented with 25 and 30 percent rice bran yielded the highest BE (98.42 and 99.84 percent, respectively, about 230 days after spawning) and MNM and initially produced a lower MMW than the substrates supplemented with 15 and 20 percent rice bran. Any amount of rice bran added to the sugarcane bagasse improved mushroom quality, with the best production of marketable mushrooms obtained by the addition of 15 percent rice bran. The largest amount of sugarcane molasses (60 g/kg) increased BE (90.3 and 23.6 percent, on first and second accumulated productions, respectively) and MNM and no quantity affected mushroom quality.