Skip to main content
Premium Trial:

Request an Annual Quote

Noncoding Structural Variants Highlighted in Integrated Cancer Genome Analysis

NEW YORK (GenomeWeb) – An international team led by investigators at the Salk Institute for Biological Studies, Pennsylvania State University, and elsewhere has developed an integrative strategy for systematically profiling structural variants (SVs) in cancer genomes — an approach that pointed to non-coding SVs as potentially "underappreciated mutational drivers" in cancer.

The researchers took a crack at several methods for profiling SVs — including high-throughput chromosome conformation capture (Hi-C) in combination with a new Hi-C-base algorithm, whole-genome sequencing, and Bionano Genomics' Irys next-generation optical mapping — in a broad set of cancer and normal samples and cell lines. They noted that each method made it possible to pick up SVs at different scales and resolutions in a preliminary analysis on new and published data for dozens of cancer cell lines.

"Each method by itself can only review a portion of the structural variations, but when you integrate the results of the three different methods, you can get the most comprehensive view of the cancer genome," co-corresponding author Feng Yue, a biochemistry and molecular biology researcher at Penn State, said in a statement. 

By bringing the Hi-C, sequencing, and optical mapping data together with a new computational pipeline, the team identified diverse structural variant patterns that might otherwise be missed in the cancer genomes — results they reported online today in Nature Genetics. These included complex structural variants and structural changes that co-occur in the same haplotype.

Through extensive Hi-C profiling, in particular, they uncovered frequent between-chromosome interactions that were missing in normal cell lines, got clues to replication timing based on translocation patterns, and began teasing apart the regulatory consequences of SVs, including those that influenced DNA folding to produce new "topologically associating domains," or TADs.

"We identified numerous instances of three-dimensional genome organization alterations as a result of structural genome variation, such as the formation or dissolution of topologically associating domains, suggesting a critical role for structural variation in gene misregulation in oncogenesis," Yue and his co-authors wrote. 

In 20 cancer cell lines, for example, the team's Hi-C analyses uncovered so-called "neo-TADs" that encompassed cancer driver genes such as ERBB2 or TERT. Moreover, gene expression data from several cancer cell lines highlighted expression shifts suspected of stemming from regulatory changes that were introduced when the genome's three-dimensional structure shifted in response to SVs.

"Determining whether any individual neo-TAD represents a recurrent alteration in a given cancer cell type, or how neo-TADs may ultimately contribute to oncogenesis, remains to be elucidated," the authors noted. "However, our analysis suggests that creation of neo-TADs is a common consequence of rearrangements in cancer genomes."

Even so, the researchers emphasized that their Hi-C-focused approach has limited power in detecting alterations less than one megabase in size, while optical mapping excels at detecting complex SVs and resolving local genome structure, but can't detect small deletions and insertions.

Whole-genome sequencing "has the highest resolution in detecting structural variation, but is less successful in detecting SVs in poorly mappable regions of the genome or in resolving complex SVs," they wrote, underlining the importance of integrative SV analyses.