Training of Deep Learning Pipelines on Memory-Constrained GPUs via Segmented Fused-Tiled Execution.

Training models with massive inputs is a significant challenge in the development of Deep Learning pipelines to process very large digital image datasets as required by Whole Slide Imaging (WSI) in computational pathology and analysis of brain fMRI images in computational neuroscience. Graphics Processing Units (GPUs) represent the primary workhorse in training and inference of Deep Learning models. In order to use GPUs to run inference or training on a neural network pipeline, state-of-the-art machine learning frameworks like PyTorch and TensorFlow currently require that the collective memory on the GPUs must be larger than the size of the activations at any stage in the pipeline.

Rapid evaporative ionization mass spectrometry (REIMS) combined with chemometrics for real-time beer analysis.

The beer industry plays an important role in the economy since this is the third most consumed beverage worldwide. Efficient analytical methods must be developed to ensure the quality of the product. Rapid evaporative ionization mass spectrometry (REIMS) can provide molecular-level information, while enabling fast analysis.

Role of epigenetics in unicellular to multicellular transition in Dictyostelium.

Background: The evolution of multicellularity is a critical event that remains incompletely understood. We use the social amoeba, Dictyostelium discoideum, one of the rare organisms that readily transits back and forth between both unicellular and multicellular stages, to examine the role of epigenetics in regulating multicellularity.

Results: While transitioning to multicellular states, patterns of H3K4 methylation and H3K27 acetylation significantly change.

Analysis of chemosensory markers in cigarette smoke from different tobacco varieties by GC×GC-TOFMS and chemometrics.

Commercial cigarettes are made from a blend of different tobacco varieties, which in turn are the results of different agronomic practices and post-harvest curing processes. The highly complex mixture of smoke compounds reflects each tobacco variety and the levels of sensory-relevant markers. Therefore, the aim of this work was to identify potential relevant chemosensory markers in the mainstream smoke of four main types of commercial tobaccos and establish any possible relationship between them and the tobacco growing/curing practices.