A bibliometric literature review of stock price forecasting: From statistical model to deep learning approach.

We introduce a comprehensive analysis of several approaches used in stock price forecasting, including statistical, machine learning, and deep learning models. The advantages and limitations of these models are discussed to provide an insight into stock price forecasting. Traditional statistical methods, such as the autoregressive integrated moving average and its variants, are recognized for their efficiency, but they also have some limitations in addressing non-linear problems and providing long-term forecasts. Machine learning approaches, including algorithms such as artificial neural networks and random forests, are praised for their ability to grasp non-linear information without depending on stochastic data or economic theory. Moreover, deep learning approaches, such as convolutional neural networks and recurrent neural networks, can deal with complex patterns in stock prices. Additionally, this study further investigates hybrid models, combining various approaches to explore their strengths and counterbalance individual weaknesses, thereby enhancing predictive accuracy. By presenting a detailed review of various studies and methods, this study illuminates the direction of stock price forecasting and highlights potential approaches for further studies refining the stock price forecasting models.

Liquid chromatography-tandem mass spectrometry analysis of peptides separated from the insoluble matrix by in-sample tryptic protein digestion for rapid discrimination of various induced pathological states of human bone models in oral surgery.

For the understanding of pathological states of bone tissues in oral surgery, it would be desirable to have the possibility to simulate these processes on bone cell models in vitro. These cultures, similarly to bone tissues, contain numerous proteins entrapped in the insoluble matrix. The major goal of this study was to verify whether a method based on direct in-matrix protein digestion could be suitable for the discrimination between different induced pathological states of bone cell models cultivated in vitro. Using in-sample specific protein digestion with trypsin followed by liquid chromatography-tandem mass spectrometry analysis of released peptides, 446 proteins (in average per sample) were identified in a bone cell in vitro model with induced cancer, 440 proteins were found in a model with induced inflammation, 451 proteins were detected in control in vitro culture, and 491 proteins were distinguished in samples of vestibular laminas of maxillary bone tissues originating from six different patients. Subsequent partial least squares - discrimination analysis of obtained liquid chromatography-tandem mass spectrometry data was able to discriminate among in vitro cultures with induced cancer, with induced inflammation, and control cultivation. Thus, the direct in-sample protein digestion by trypsin followed by liquid chromatography-tandem mass spectrometry analysis of released specific peptide fragments from the insoluble matrix and mathematical analysis of the mass spectrometry data seems to be a promising tool for the routine proteomic characterization of in vitro human bone models with induced different pathological states.

A small molecule UPR modulator for diabetes identified by high throughput screening

Unfolded protein response (UPR) is a stress response that is specific to the endoplasmic reticulum (ER). UPR is activated upon accumulation of unfolded (or misfolded) proteins in the ER's lumen to restore protein folding capacity by increasing the synthesis of chaperones. In addition, UPR also enhances degradation of unfolded proteins and reduces global protein synthesis to alleviate additional accumulation of unfolded proteins in the ER. Herein, we describe a cell-based ultra-high throughput screening (uHTS) campaign that identifies a small molecule that can modulate UPR and ER stress in cellular and