DECEMBER 19TH, 2019
The shortest bi-weekly genomics research report in the world. Curated by the Emedgene Research Division.
Is AI making healthcare more human?
MIT Technology Review and GE Healthcare surveyed 900 healthcare professionals on AI within their organizations. Over 70% believe AI improves workflows and disease predictions, while also preventing burn out. Over 80% of admins believe that AI makes them more competitive.
MIT Technology Review
“AI won’t replace radiologists, but radiologists who use AI will replace radiologists who don’t”. The winning quote from a Nature news piece titled ‘The Rise of Robot Radiologists’.
Moving AI from research to clinical use
The National Academy of Medicine issued a new report: ‘AI in Healthcare: The Hope, The Hype, The Promise, The Peril’. With such a great title, we couldn’t help but read. They cover the key issues we need to address in AI in order to bridge the gap from research to practical use, diversity, overcoming bias, transparency, augmenting rather than replacing humans and more.
In an example of bridging the gap between research and implementation in clinical care, Mayo Clinic’s AI team is located in the hospital. As one of the co-directors puts it: “You learn to build the AI in a way that fits the physician practice, and not the other way around”
Our models are continuously learning and improving. How do we handle this from a regulatory perspective? Should we lockdown our algorithms once the FDA has approved them?
What is XAI and why do we need it?
The purpose of an explainable AI (XAI) system is to make its behavior more intelligible to humans by providing explanations, making the black box slightly more transparent. Among the XAI challenges raised in this review is the question of providing explanations for model features or decisions.
Should we demand explainable AI in healthcare? Or is ‘black box’ medicine acceptable?
Annals of Internal Medicine
Just as a reminder, Emedgene’s AI system is a good example of XAI, and believe explaining the model’s decision can help geneticists utilize AI more effectively.
Robot Break !
Even robots foster trust by explaining their actions Science Robotics
Meanwhile, in the clinical genome,
Small open reading frames (smORFs) are the emerging class of genes that might play a critical role in several biological mechanisms. Saghatelian et al integrated de novo transcriptome assembly and Ribo-seq for the accurate annotation of these tiny proteins and identified >2500 protein-coding smORFs.
Nature Chemical Biology
Pseudogenes have been frequently dismissed from functional assessment and genomic analysis due to their misleading label. However, recent studies indicated their crucial role in several biological processes. The current review emphasizes on the reassessment of these genomic elements to enhance our understanding of genome function and evolution.
Nature Review Genetics
Prioritizing genes in de novo SVs. Researchers combined the analyses of WGS data with phenotype association and chromatin organization datasets to improve diagnosis for patients with previously inconclusive results.
Identification of unique transcriptional machinery required by GC-rich repeat expansions may highlight novel therapeutic targets universal to repeat-expansion associated diseases.
Trends in Genetics
End of year highlights, and predictions
The Genomic Medicine Working Group of the National Advisory Council for Human Genome Research of the NHGRI chose the top 10 accomplishments in the field published in 2019.
Highlights from a year of Nature Genetics
What will the next 25 years in medicine look like? Improving quality and affordability.
Expanding DNA screening, responsibly
Can we move away from medical history–based screening and toward optimized DNA-based population screening? Many questions need to be answered before commencing with routine population-based screening.
And if we do perform screening, should doctors have a legal duty to warn relatives of their genetic risks?
Therapies for rare diseases
Despite major progress in the rare disease knowledge, only a few of them have approved treatments. This comprehensive review by Prunotto & co-authors throws light on the major advancements and challenges in the treatment of rare genetic disorders.
Nature Reviews Drug Discovery
- MN1 and craniofacial and brain abnormalities AJHG
- XRCC6 and autism spectrum disorder Journal of Human Genetics
- RHOA and novel neuro‐ectodermal manifestations Human Mutation
- SETD1A and neurodevelopmental disorder bioRxiv
- FBN1 and congenital scoliosis Journal of Human Genetics
(Ada is our AI engine, she likes to add an interesting new factoid she discovered in every issue)
Biallelic pathogenic variants in DONSON have been identified as a novel cause of Meier-Gorlin syndrome. BMJ Journals
Questions? Comments? Corrections? Please email us at email@example.com