gaining a new function. One of the first cancer-promoting genetic alterations discovered
was the BCR–ABL gene fusion, also referred to as the Philadelphia chromosome, in
leukemia cells in 1959. Since then, more than 300 genes have been classified as
participating in driver fusion events by either gaining oncogenic potential or by modifying
their partner’s function [1].
Review Question 3
When analyzing cancer RNA-Seq data, how would you identify gene fusions?
2.3
Sequencing in Cancer Diagnosis
Knowing the genes that are involved in cancer development, coupled with the ability to
cost-effectively perform gene sequencing and bioinformatic analyses, is a powerful tool to
screen patients at risk and aid genetic counseling.
At the moment, genetic tests can be requested by individuals from cancer-prone families
to learn whether they are also at a higher risk for developing the disease. If the family
carries a known mutation in a cancer gene, it can be identified through gene panels, which
usually sequence a small number of known genes by hybridization followed by highthroughput sequencing (targeted sequencing) or by PCR followed by capillary sequencing.
In the event that the gene panel testing returns with a negative result, the family may enter a
research protocol where whole-exome or genome sequencing will be performed to attempt
to identify novel cancer genes. These projects are usually research-focused (i.e., no
information is returned to the patient) and can increase their statistical detection power
by aggregating a large number of families. However, bioinformatic analysis is key and both
of these methodologies suffer from the identification of a large number of variants of
uncertain significance (VUS).
VUS represent a challenge for bioinformaticians, medical professionals, and patients
alike because their relationship to disease risk is unknown and therefore clinically
unactionable. In order to alleviate this issue, the American College of Medical Genetics
and Genomics (ACMG) published in 2000, and revised in 2007 and 2015 [6], a series of
recommendations to classify variants into five categories based on population frequencies,
computational predictions, functional data, and familial disease/variant co-segregation
observations.
Variant classification:
• pathogenic
• likely pathogenic
• uncertain significance
• likely benign
• benign
2 Opportunities and Perspectives of NGS Applications in Cancer Research
21
was the BCR–ABL gene fusion, also referred to as the Philadelphia chromosome, in
leukemia cells in 1959. Since then, more than 300 genes have been classified as
participating in driver fusion events by either gaining oncogenic potential or by modifying
their partner’s function [1].
Review Question 3
When analyzing cancer RNA-Seq data, how would you identify gene fusions?
2.3
Sequencing in Cancer Diagnosis
Knowing the genes that are involved in cancer development, coupled with the ability to
cost-effectively perform gene sequencing and bioinformatic analyses, is a powerful tool to
screen patients at risk and aid genetic counseling.
At the moment, genetic tests can be requested by individuals from cancer-prone families
to learn whether they are also at a higher risk for developing the disease. If the family
carries a known mutation in a cancer gene, it can be identified through gene panels, which
usually sequence a small number of known genes by hybridization followed by highthroughput sequencing (targeted sequencing) or by PCR followed by capillary sequencing.
In the event that the gene panel testing returns with a negative result, the family may enter a
research protocol where whole-exome or genome sequencing will be performed to attempt
to identify novel cancer genes. These projects are usually research-focused (i.e., no
information is returned to the patient) and can increase their statistical detection power
by aggregating a large number of families. However, bioinformatic analysis is key and both
of these methodologies suffer from the identification of a large number of variants of
uncertain significance (VUS).
VUS represent a challenge for bioinformaticians, medical professionals, and patients
alike because their relationship to disease risk is unknown and therefore clinically
unactionable. In order to alleviate this issue, the American College of Medical Genetics
and Genomics (ACMG) published in 2000, and revised in 2007 and 2015 [6], a series of
recommendations to classify variants into five categories based on population frequencies,
computational predictions, functional data, and familial disease/variant co-segregation
observations.
Variant classification:
• pathogenic
• likely pathogenic
• uncertain significance
• likely benign
• benign
2 Opportunities and Perspectives of NGS Applications in Cancer Research
21