The digital revolution in phenotyping

Abstract

Phenotypes have gained increased notoriety in the clinical and biological domain owing to their application in numerous areas such as the discovery of disease genes and drug targets, phylogenetics and pharmacogenomics. Phenotypes, defined as observable characteristics of organisms, can be seen as one of the bridges that lead to a translation of experimental findings into clinical applications and thereby support ‘bench to bedside’ efforts. However, to build this translational bridge, a common and universal understanding of phenotypes is required that goes beyond domain-specific definitions. To achieve this ambitious goal, a digital revolution is ongoing that enables the encoding of data in computer-readable formats and the data storage in specialized repositories, ready for integration, enabling translational research. While phenome research is an ongoing endeavor, the true potential hidden in the currently available data still needs to be unlocked, offering exciting opportunities for the forthcoming years. Here, we provide insights into the state-of-the-art in digital phenotyping, by means of representing, acquiring and analyzing phenotype data. In addition, we provide visions of this field for future research work that could enable better applications of phenotype data.

Benchmarking deep learning models on large healthcare datasets

“…Physionet challenge 1a friendly competition platform -has resulted in development of machine learning models for addressing some of the open healthcare problems. With the recent advances in deep learning techniques, there is a growing interest in applying these techniques to healthcare applications due to the increasing availability of large-scale health care data [34,7,35,36]. For example, Che et al [7] developed a scalable deep learning framework which models the prior-knowledge from medical ontologies to learn clinically relevant features for disease diagnosis.…”

BioHackathon 2015: Semantics of data for life sciences and reproducible research

“…In addition, while phenotypes in the biological domain are recorded as results from biological experiments, phenotypes in the clinical domain are used to report the state condition of patients 186 . Furthermore, in current clinical nomenclatures for phenotypes such as MeSH, the 10th revision of the International Statistical Classification of Diseases and Related Health Problems (ICD-10), the nomenclature of the National Cancer Institute (NCI), SNOMED Clinical Terms (SNOMED CT), and UMLS, concepts are covered inconsistently and incompletely 186 . All these issues affect ontology interoperability, and thus, the quality of their applications.…”

Early‐onset inflammatory bowel disease as a model disease to identify key regulators of immune homeostasis mechanisms

“…In addition to providing a unified nomenclature that allows clinicians and researchers to characterize patients better, the inherent network structure of ontologies such as HPO and GO allows for pairwise distance (similarity) between two terms. Consequently, pairwise similarity measures can be used to carry out complex comparisons such as testing the similarity between two patients annotated by different terms . Current ontologies have been useful in fulfilling current gap, but they are not complete.…”

“…Consequently, pairwise similarity measures can be used to carry out complex comparisons such as testing the similarity between two patients annotated by different terms. 251 Current ontologies have been useful in fulfilling current gap, but they are not complete. Numerous diseases and diseasegene association are not documented, and the ontology structures are incomplete.…”