AI Helps Scientists Examine Gene Activation

September 17, 2020 10:56:59

Scientists from UC, San Diego have finally worked out an enigma in human gene activation. As illustrated in the journal Nature, this discovery may be used to control gene activation in biomedical and biotechnology applications.

For a long time, scientists have known that the human genes act using instructions that are delivered by the specific order of a human’s DNA. This is directed by the 4 types of individual links, coded C, A, T and G.

About 25% of a human’s genes are transcribed in sequences resembling TATAAA, which is referred to as the TATA box. However, for the remaining 75%, it’s still a mystery how the genes are turned on or promoted, given the huge number of DNA link sequence possibilities.

Using artificial intelligence, researchers at UC San Diego have discovered a DNA activation code that is used almost as often as the TATA box in human beings. This discovery, which the scientists termed as the DPR (downstream core promoter region), may be used in the future to control gene activation.

A distinguished professor from UC San Diego’s Division of Biological Sciences and the senior author of the paper, James T. Kadonaga, stated that the DPR identification shows an important stage in the activation of roughly 75% of a human’s genes. He also credited the discovery of this new information to the use of machine learning.

About 3 decades ago in 1996, Kadonaga together with his colleagues discovered a gene activation sequence while they were working in fruit flies, which they named the DPE. It enables the turning on of genes in the absence of the TATA box. Since that time though, the research has stalled in terms of deciphering additional details concerning the human DPE.

For this study, Kadonaga collaborated with Cassidy Yunjing Huang, post-doctoral scholar and lead author, Long Vo Ngoc, Claudia Medrano and Jack Cassidy, who is a retired computer scientist that assisted the team in acquiring the necessary tools of artificial intelligence.

The study results showed the existence of the Downstream core Promoter Region motif in human genes. Additionally, the occurrence frequency of the DPR is comparable to the TATA box one. They discovered that the genes that were activated with the TATA box sequences lacked the DPR sequences.

In the future, the use of artificial intelligence to analyze DNA sequence patterns will increase scientists’ ability to study and understand and also control gene activation in human cells. Analysts say this technology could open up new applications for biomed companies like Predictive Oncology (NASDAQ: POAI).

About BioMedWire

BioMedWire (BMW) is a bio-med news and content distribution company that provides (1) access to a network of wire services via NetworkWire to reach all target markets, industries and demographics in the most effective manner possible, (2) article and editorial syndication to 5,000+ news outlets (3), enhanced press release services to ensure maximum impact, (4) social media distribution via the Investor Brand Network (IBN) to nearly 2 million followers, (5) a full array of corporate communications solutions, and (6) a total news coverage solution with BMW Prime. As a multifaceted organization with an extensive team of contributing journalists and writers, BMW is uniquely positioned to best serve private and public companies that desire to reach a wide audience of investors, consumers, journalists and the general public. By cutting through the overload of information in today’s market, BMW brings its clients unparalleled visibility, recognition and brand awareness. BMW is where news, content and information converge.

To receive SMS text alerts from BioMedWire, text “STOCKS” to 77948 (U.S. Mobile Phones Only)

For more information, please visit

Please see full terms of use and disclaimers on the BioMedWire website applicable to all content provided by BMW, wherever published or re-published:

BioMedWire (BMW)
San Francisco, California
415.949.5050 Office

BioMedWire is part of the InvestorBrandNetwork.