New computational tool harnesses big data, deep learning to reveal dark matter of the transcriptome

March 25, 2019

A research team at Children's Hospital of Philadelphia (CHOP) has developed an innovative computational tool offering researchers an efficient method for detecting the different ways RNA is pieced together (spliced) when copied from DNA. Because variations in how RNA is spliced play crucial roles in many diseases, this new analytical tool will provide greater capabilities for discovering disease biomarkers and therapeutic targets, even from RNA-sequencing data sets with modest coverage.

Study leader Yi Xing, PhD, director of the Center for Computational and Genomic Medicine at CHOP, and first authors and PhD students Zijun Zhang and Zhicheng Pan report on their DARTS framework this week in Nature Methods. DARTS (Deep-learning Augmented RNA-seq analysis of Transcript Splicing) uses deep-learning based predictions to harness the wealth of information available in public datasets of RNA sequencing (RNA-seq), thus allowing for new insights into alternative splicing.

"The conceptual innovation of DARTS is it provides a bridge from big data in the public domain to smaller data sets in focused studies with individual investigators," said Xing. "DARTS offers the ability to transform massive amounts of public RNA-seq data into a knowledge base, represented as a deep neural network, of how splicing is regulated. Using this computational framework, we can push that into any individual lab. This could be really useful and increase the efficiency of the experiment and enable new discoveries. With just 20 or 30 million RNA-seq reads, you can make educated guesses and inferences on things you were never able to see in the past."

Xing has a long-standing research focus on alternative splicing--the process by which information in DNA of a single gene is pieced together in different ways to generate different messenger RNA and protein products after gene transcription. Genes each generate an average of 10 or more such products, and sometimes as many as 38,000. Those variations in alternative splicing may cause disease, modify disease risk, or make a disease milder or worse.

Massively parallel RNA sequencing is now the standard technology researchers use to investigate alternative splicing. However, to accurately measure alternative splicing, the RNA sequencing experiments have to go very deep. The consensus view is that over 100 million sequences are needed for analyzing alternative splicing, but due to the high cost, most researchers cannot afford going this deep with their RNA sequencing experiments. Moreover, many medically important genes are not expressed at high levels. Even a deep RNA sequencing experiment cannot generate enough coverage on such genes, making it virtually impossible to measure the genes' alternative splicing patterns.

In the current study, Xing's team first drew on large-scale public-domain RNA sequencing data from sources such as the ENCODE Consortium, the international program launched by the National Human Genome Research Institute, to identify all the functional elements in the genome, including those acting at the level of RNA. Using these massive data sets, DARTS trains a deep neural network for predicting changes in alternative splicing. The model incorporates messenger RNA (mRNA) levels of 1,500 RNA binding proteins and 3,000 sequence features.

To allow researchers to use the deep learning model in their own studies, the deep neural network predictions are combined with actual RNA sequencing data generated on specific biological samples using a statistical framework called Bayesian hypothesis testing. Researchers can use this information in their individual labs to better characterize alternative splicing across different biological conditions.

The researchers applied DARTS to lung and prostate cancer cell lines to test its ability to predict splicing patterns in the cells. These cell lines are models for the transition from epithelial to mesenchymal cells--an important process in both embryonic development and cancer metastasis. By leveraging the deep learning predictions, DARTS discovered changes in alternative splicing patterns in numerous genes that escaped detection by conventional computational tools because these genes were expressed at low levels in the cells. The study team then performed experiments to validate these novel predictions. These new discoveries may allow scientists to better identify biomarkers and therapeutic targets of diseases.

"DARTS offers an exciting conceptual framework that we could adapt to other uses," added Xing. "For example, we might create a version that predicts alternative splicing in specific patient tissues." This could potentially improve diagnosis of rare diseases from a tissue biopsy, a useful technique for pediatric centers such as CHOP that often evaluate children with puzzling, undiagnosed disorders.

DARTS, Xing concluded, could enable scientists to discover more about the contributions of understudied genes that may not be expressed at high levels, but have important impacts on health and disease. "DARTS offers a new window into the dark matter of the transcriptome," he said.
-end-
DARTS is broadly available to the scientific community at https://github.com/Xinglab/DARTS.

The National Institutes of Health provided funding support for this study (grants R01GM088342, R01GM117624, U01HG007912 and U01CA233074). Disclosure: Xing and study co-author Douglas L. Black of the University of California, Los Angeles, are scientific co-founders of the company Panorama Medicine. Panorama holds a non-exclusive license to the DARTS technology. Xing and co-author Zijun Zhang have filed a provisional patent application for DARTS.

Zijun Zhang et al, "Deep-learning augmented RNA-seq analysis of transcript splicing," Nature Methods, March 25, 2019. https://doi.org/10.1038/s41592-019-0351-9

About Children's Hospital of Philadelphia: Children's Hospital of Philadelphia was founded in 1855 as the nation's first pediatric hospital. Through its long-standing commitment to providing exceptional patient care, training new generations of pediatric healthcare professionals, and pioneering major research initiatives, Children's Hospital has fostered many discoveries that have benefited children worldwide. Its pediatric research program is among the largest in the country. In addition, its unique family-centered care and public service programs have brought the 564-bed hospital recognition as a leading advocate for children and adolescents. For more information, visit http://www.chop.edu

Children's Hospital of Philadelphia

Related Dark Matter Articles from Brightsurf:

Dark matter from the depths of the universe
Cataclysmic astrophysical events such as black hole mergers could release energy in unexpected forms.

Seeing dark matter in a new light
A small team of astronomers have found a new way to 'see' the elusive dark matter haloes that surround galaxies, with a new technique 10 times more precise than the previous-best method.

Holding up a mirror to a dark matter discrepancy
The universe's funhouse mirrors are revealing a difference between how dark matter behaves in theory and how it appears to act in reality.

Zooming in on dark matter
Cosmologists have zoomed in on the smallest clumps of dark matter in a virtual universe - which could help us to find the real thing in space.

Looking for dark matter with the universe's coldest material
A study in PRL reports on how researchers at ICFO have built a spinor BEC comagnetometer, an instrument for studying the axion, a hypothetical particle that may explain the mystery of dark matter.

Looking for dark matter
Dark matter is thought to exist as 'clumps' of tiny particles that pass through the earth, temporarily perturbing some fundamental constants.

New technique looks for dark matter traces in dark places
A new study by scientists at Lawrence Berkeley National Laboratory, UC Berkeley, and the University of Michigan -- published today in the journal Science - concludes that a possible dark matter-related explanation for a mysterious light signature in space is largely ruled out.

Researchers look for dark matter close to home
Eighty-five percent of the universe is composed of dark matter, but we don't know what, exactly, it is.

Galaxy formation simulated without dark matter
For the first time, researchers from the universities of Bonn and Strasbourg have simulated the formation of galaxies in a universe without dark matter.

Taking the temperature of dark matter
Warm, cold, just right? Physicists at UC Davis are using gravitational lensing to take the temperature of dark matter, the mysterious substance that makes up about a quarter of our universe.

Read More: Dark Matter News and Dark Matter Current Events
Brightsurf.com is a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to Amazon.com.