AI Machine Learning & Data Science Research

Scientists Use DL and Other Tools for 2019 Novel Coronavirus Host and Infectivity Prediction

A new study suggests human-to-human transmission of the 2019 Novel Coronavirus (2019-nCoV) may have started as early as mid December, 2019.

A new study suggests human-to-human transmission of the 2019 Novel Coronavirus (2019-nCoV) may have started as early as mid December, 2019. The findings contradict Wuhan Health Commission statements that “no significant evidence of human-to-human transmission was found” for the 2019-nCoV in late December or early January.

The research was published yesterday in the prestigious peer-reviewed medical journal The New England Journal of Medicine (NEJM), by researchers from Chinese Center for Disease Control and Prevention, the centers for disease control and prevention in various provinces and cities in China, and several research institutes. The NEJM stresses that “views expressed in this article are those of the authors and do not represent the official policy of the China CDC.”

The initial cases of the recent novel coronavirus–infected pneumonia occurred in the Huanan Seafood Wholesale Market of Wuhan, Hubei Province, China, in December 2019. Researchers analyzed data on the first 425 confirmed Wuhan cases, and estimated the basic reproductive number of the 2019-nCoV to be around 2.2.

Basic reproductive number or R0 represents the number of cases one case generates on average in an uninfected population over the course of its infectious period. An R0 greater than 1 means the infection will be able to spread in a population.

bn-130.webp

The researchers say efforts to control the outbreak should include close monitoring to identify effective measures to reduce transmission in the community, and watching for any changes in epidemiology — for example, increases in infections among persons in younger age groups or health care workers.

A number of other research teams in China and worldwide have published papers analyzing the spread of the outbreak, with some applying machine learning algorithms to assist in their analysis.

Last week, researchers from Peking University and the First Affiliated Hospital of Zhejiang University’s College of Medicine published Host and Infectivity Prediction of Wuhan 2019 Novel Coronavirus Using Deep Learning Algorithm on bioRxiv, which suggests that 2019-nCoV infectivity patterns closely resemble 2003’s severe acute respiratory syndrome coronavirus (SARS-CoV), other Bat SARS-like Coronaviruses WIV1, and the Middle East respiratory syndrome coronavirus (MERS-CoV) of 2012.

Because the 2019-nCoV infectivity patterns also closely resemble those of mink viruses, researchers identified bats and minks as two main candidate reservoirs for the 2019-nCoV.

Studies have come up with a range of 2019-nCov R0 estimations. The Guangdong Provincial Center for Disease Control and Prevention in China estimated a R0 of 2.9 using Exponential Growth and maximum likelihood estimation, a Hong Kong University team predicted an R0 of 2.13, and the most recent Imperial College London R0 estimate is 2.6.

Additional studies are underway to further improve 2019-nCoV R0 and host prediction. Synced is watching this topic closely and will continue to update readers as new information becomes available.


Journalist: Yuan Yuan | Editor: Michael Sarazen

3 comments on “Scientists Use DL and Other Tools for 2019 Novel Coronavirus Host and Infectivity Prediction

  1. Thank you for the topic.

  2. Thanks for all the research paper it was very informative

  3. Thank you.

Leave a Reply to zoli zoli Cancel reply

Your email address will not be published.

%d bloggers like this: