Deep Learning for Human Disease Detection, Subtype Classification, and Treatment Response Prediction Using Epigenomic Data

P.O.L.A.R. Star: A New Framework Developed and Applied by One Mid-Sized Pharmaceutical Company to Drive Digital Transformation in R&D

Digital transformation has become a cornerstone of innovation in pharmaceutical research and development (R&D). Pharmaceutical companies now have an imperative to embrace transformation, including mid-sized and small-sized companies despite resource limitations that do not allow economies of scale compared with larger organizations. This article describes the journey undertaken by Chiesi to develop an efficient framework to drive digital transformation along its R&D value chain with the objective of building and refreshing a clear roadmap and relevant priorities, together with identifying and enabling new digital capabilities and skills within R&D, defining tools and processes that will guide Chiesi activities in the space up to mid-long term. This work has led so far to five main achievements, which align with the steps in the framework: a strategically aligned roadmap with key focus areas for digital transformation and a dedicated team to lead the effort; a common language for data across the R&D value chain; an internal mindset that's open to innovation and participation in key external networks and consortia; a set of quick-win use cases for the new framework; and a defined set of Key Performance Indicators (KPIs) and monitoring tools for digital transformation. The work presented here demonstrates that R&D digital transformation should represent an ongoing process to enable cross-functional collaboration and integration within complex corporate environments that face an ever-growing volume of diverse data, to efficiently support business needs, and to ensure a positive impact on patient care.

A Novel Image Processing Method for Obtaining an Accurate Three-Dimensional Profile of Red Blood Cells in Digital Holographic Microscopy

Recently, research on disease diagnosis using red blood cells (RBCs) has been active due to the advantage that it is possible to diagnose many diseases with a drop of blood in a short time. Representatively, there are disease diagnosis technologies that utilize deep learning techniques and digital holographic microscope (DHM) techniques. However, three-dimensional (3D) profile obtained by DHM has a problem of random noise caused by the overlapping DC spectrum and sideband in the Fourier domain, which has the probability of misjudging diseases in deep learning technology. To reduce random noise and obtain a more accurate 3D profile, in this paper, we propose a novel image processing method which randomly selects the center of the high-frequency sideband (RaCoHS) in the Fourier domain. This proposed algorithm has the advantage of filtering while using only recorded hologram information to maintain high-frequency information. We compared and analyzed the conventional filtering method and the general image processing method to verify the effectiveness of the proposed method. In addition, the proposed image processing algorithm can be applied to all digital holography technologies including DHM, and in particular, it is expected to have a great effect on the accuracy of disease diagnosis technologies using DHM.

Machine Learning Methods for Cancer Classification Using Gene Expression Data: A Review

Cancer is a term that denotes a group of diseases caused by the abnormal growth of cells that can spread in different parts of the body. According to the World Health Organization (WHO), cancer is the second major cause of death after cardiovascular diseases. Gene expression can play a fundamental role in the early detection of cancer, as it is indicative of the biochemical processes in tissue and cells, as well as the genetic characteristics of an organism. Deoxyribonucleic acid (DNA) microarrays and ribonucleic acid (RNA)-sequencing methods for gene expression data allow quantifying the expression levels of genes and produce valuable data for computational analysis. This study reviews recent progress in gene expression analysis for cancer classification using machine learning methods. Both conventional and deep learning-based approaches are reviewed, with an emphasis on the application of deep learning models due to their comparative advantages for identifying gene patterns that are distinctive for various types of cancers. Relevant works that employ the most commonly used deep neural network architectures are covered, including multi-layer perceptrons, as well as convolutional, recurrent, graph, and transformer networks. This survey also presents an overview of the data collection methods for gene expression analysis and lists important datasets that are commonly used for supervised machine learning for this task. Furthermore, we review pertinent techniques for feature engineering and data preprocessing that are typically used to handle the high dimensionality of gene expression data, caused by a large number of genes present in data samples. The paper concludes with a discussion of future research directions for machine learning-based gene expression analysis for cancer classification.

Prediction of Gastric Cancer-Related Genes Based on the Graph Transformer Network

Gastric cancer is a complex multifactorial and multistage process that involves a large number of tumor-related gene structural changes and abnormal expression. Therefore, knowing the related genes of gastric cancer can further understand the pathogenesis of gastric cancer and provide guidance for the development of targeted drugs. Traditional methods to discover gastric cancer-related genes based on biological experiments are time-consuming and expensive. In recent years, a large number of computational methods have been developed to identify gastric cancer-related genes. In addition, a large number of experiments show that establishing a biological network to identify disease-related genes has higher accuracy than ordinary methods. However, most of the current computing methods focus on the processing of homogeneous networks, and do not have the ability to encode heterogeneous networks. In this paper, we built a heterogeneous network using a disease similarity network and a gene interaction network. We implemented the graph transformer network (GTN) to encode this heterogeneous network. Meanwhile, the deep belief network (DBN) was applied to reduce the dimension of features. We call this method “DBN-GTN”, and it performed best among four traditional methods and five similar methods.