Biomedical Engineering Reference
In-Depth Information
all variants in each patient. An automatic fi ltering approach must be applied for
WES/WGS studies which may even need to be considered for large disease-targeted
panels. Therefore, it is important to understand the variant fi ltering strategy to
understand the sensitivity and limitations of test results. In addition, it is well-known
that many databases contain misclassifi ed variants, particularly benign variation
misclassifi ed as disease-causing (Bell et al. 2011 ). As a result, there are very few
variant databases which are professionally curated to a clinical grade with evidence-
based assessment of clinical and function data. In addition, most Mendelian dis-
eases have a large percentage of variants that are private (unique to families)
requiring a robust process for assessing novel variation and clinical laboratories that
identifi ed these mutations may not release the information via publications in a
timely manner.
In the analysis of WES/WGS, common assumptions have to be made to set the
algorithm for variant fi ltering, which include (1) causative mutations for Mendelian
disorders are rare, (2) disease is highly penetrant, (3) mode of disease inheritance,
and (4) phenotype-based fi ltering (Majewski et al. 2011 ). Successfully identifying
the molecular basis for a rare disorder may depend on the strategy employed, such
as availability of appropriate family members for comparison, given a suspected
mode of inheritance. Regardless of the approach employed, it is recommended that
referring physicians provide detailed phenotypic information to assist the clinical
laboratory in analyzing and interpreting the results of testing. This step is a neces-
sity for WES/WGS to enable appropriate fi ltering strategies to be employed. It is
also highly recommended for large disease panel testing, given the diversity of
genes and sub-phenotypes that may be included in a test panel. The ability for clini-
cal laboratories to prioritize variants for further consideration or likely relevance
may be dependent on the constellation of symptoms and fi ndings in the patient
(phenotype fi lter). However, physicians typically have very busy schedules and the
complete clinical information may not be communicated at the time of requesting
the test. A good practice is to follow up with the physician to complete the clinical
inquiries prior to the data analyses. Clinical laboratories must set thresholds to bal-
ance over-fi ltering that could inadvertently exclude causative variants with under-
fi ltering that presents too many variants for farther analyses. Clinical laboratories
should provide a summary of the variant fi ltering strategy to assist ordering health-
care professionals to understand the sensitivity and specifi city of the test results.
9.4
Testing Validation and Revalidation
Various combinations of instruments, reagents, and analytical pipelines may be
used in tests involving NGS. The entire test should be validated using allowed sam-
ple types before clinical laboratory implementation. Assay performance character-
istics including analytical sensitivity and specifi city as well as the assay's
repeatability and reproducibility need to be established (Mattocks et al. 2010 ). For
Search WWH ::




Custom Search