Artificial Intelligence in the Genetic Diagnosis of Rare Disease
Original price was: $27.95.$2.99Current price is: $2.99.
- Title: Artificial Intelligence in the Genetic Diagnosis of Rare Disease
- Author: Kiely N. James PhD, Sujal Phadke PhD, Terence C. Wong PhD and Shimul Chowdhury PhD
- Publisher: Elsevier Inc
- Publication: 2023
- Language: English
- Pages: 17
- Format: PDF
- File size: 556 KB
- DOI: 10.1016/j.cll.2022.09.023
- ID: CBR68
Key points
- •
The use of artificial intelligence can streamline the lengthy process currently required to clinically interpret a genome.
- •
Natural language processing can eliminate much of the human variability and bias that is involved in translating information from the medical record.
- •
Deep learning approaches can expedite the prioritization of genetic variants during interpretation of genomic data.
- •
The promise of genomic medicine will be fully realized only if the time and cost of analyzing genomic data continue to decrease.
Introduction
More than 7000 rare diseases have been described, with prevalence ranging from fewer than 1 in a million (eg, metachromatic leukodystrophy) to greater than 1 in 10,000 (eg, sickle cell anemia), and of these, approximately 70% are largely genetic in origin. In total, an estimated 263 to 446 million individuals are thought to be afflicted by rare diseases worldwide . The identification of causal genetic variants in these individuals enables patient-specific clinical management, referred to as genomic or precision medicine, which has the potential to improve patient survival and quality of life, and reduce health care costs ,.
Genetic diagnosis is achieved through testing of specific genes or by more comprehensive interrogation via whole exome sequencing (WES) or whole genome sequencing (WGS; Box 1 ). The typical workflow for diagnostic genomic sequencing in affected individuals involves sample collection and processing, phenotypic evaluation, genetic variant detection, genetic variant interpretation, and reporting ( Fig. 1 ). In cases in which the pretest differential diagnosis is narrow and pathognomonic features are present, a single gene or gene panel test is likely to be ordered. However, the use of clinical WES and WGS is growing more widespread as costs and turn-around-times drop. In both methods, primary phenotypes are used during genomic analysis to inform gene review and variant prioritization based on the overlap of canonical disease descriptions with the patient’s phenotype, a necessary step, as WES and WGS can produce approximately 25,000 genetic variants and upward of 4 to 5 million variants, respectively
Be the first to review “Artificial Intelligence in the Genetic Diagnosis of Rare Disease”
Related products
Researches
Cancer survival in the northwestern of Sao ˜ Paulo State, Brazil: A population-based study
Reviews
There are no reviews yet.