Sha, Y. | bioRxiv Science

bioRxiv2017

The collective influence of 1, 25-dihydroxyvitamin D3 with physiological fluid shear stress on osteoblasts

1, 25-dihydroxyvitamin D3 (1, 25 (OH)2 D3) and mechanical stimuli in physiological environment play an important role in the pathogenesis of osteoporosis. The effects of 1, 25-dihydroxyvitamin D3 alone and mechanical stimuli alone on osteoblasts have been widely investigated. This study reports the collective influences of 1, 25-dihydroxyvitamin D3 and flow shear stress (FSS) on biological functions of osteoblasts. 1, 25 (OH)2 D3 were constructed in various kinds of concentration (0, 1, 10, 100 nmmol/L), while physiological fluid shear stress (12 dynes/cm2) were produced by using a parallel-plate fluid flow system. 1, 25 (OH)2 D3 affects the responses of ROBs to FSS, including the inhibition of NO releases and cell proliferation as well as the promotion of PGE2 releases and cell differentiation. These findings provide a possible mechanism by which 1, 25(OH)2 D3 influences osteoblasts responses to FSS and may provide guidance for the selection of 1, 25(OH)2 D3 concentration and mechanical loading in order to in vitro produce functional bone tissues.

cell biology↗

bioRxiv2017

DeepDeath: Learning To Predict The Underlying Cause Of Death With Big Data

Multiple cause-of-death data provides a valuable source of information that can be used to enhance health standards by predicting health related trajectories in societies with large populations. These data are often available in large quantities across U.S. states and require Big Data techniques to uncover complex hidden patterns. We design two different classes of models suitable for large-scale analysis of mortality data, a Hadoop-based ensemble of random forests trained over N-grams, and the DeepDeath, a deep classifier based on the recurrent neural network (RNN). We apply both classes to the mortality data provided by the National Center for Health Statistics and show that while both perform significantly better than the random classifier, the deep model that utilizes long short-term memory networks (LSTMs), surpasses the N-gram based models and is capable of learning the temporal aspect of the data without a need for building ad-hoc, expert-driven features.

pathology↗