CNV Technical Standards Web Series

A multi-part web series to educate the community about the newly released ACMG/ClinGen technical standards for interpretation and reporting of constitutional copy number variants (CNVs).

We will continue to add new FAQs to this page as they arise.  Have a question that you don’t see listed here? Email us at

Web Series Logistics

Anyone is welcome to join the web series at any time.  To receive the teleconference information, please fill out this brief registration survey:  The teleconference information will display at the end of the survey; please copy this into your own calendar invite for your reference.

We need to keep track of how many participants are interested in the series in order to make sure our teleconference can accommodate the appropriate number of people.  We would also like to track metrics on who the content is reaching - how many different countries, what types of genetics professionals, etc. When you fill out the registration survey, you provide your email address, which also allows us to contact you with information related to the web series, such as letting you know when videos have been posted.

As described above, we are interested in tracking metrics on who our content is reaching.  Additionally, we would like to get a sense of how many people attended the live teleconferences vs. watched the video, so filling out the attendance survey is important regardless of how you accessed the content.  For those individuals participating in the optional pre-/post-CNV evaluation project, it is important for us to know which sessions you attended, as we will use this as a variable when we evaluate the overall performance results.

Due to the large number of people registered for the series, all attendees will be muted upon entry to the teleconference.  To ask a question, please type it into the teleconference Q&A function. A monitor will be reviewing the Q&A throughout each call.  If your question is easily answerable via chat, the monitor will do so. Other questions will be read aloud at the end of the session as time allows.  Questions that are not addressed during a particular session will be saved for the special Question & Answer session on March 12, 2020. If you think of a question you would like to ask outside of a session, please email it to

Recordings will be made available within a few days of each session.  These recordings will be posted on both the ClinGen website and the ClinGen YouTube channel.  If you subscribe to the ClinGen YouTube channel, you will be notified when a new video is posted.  We will also announce the availability of new recordings via Twitter (@ClinGenResource). Finally, we will send reminder emails at the beginning of each week reminding you of the coming Thursday’s topic and announcing the availability of the previous Thursday’s recording.

No.  In an effort to bring you this web series as quickly as possible after the release of the technical standards, we have not applied for continuing education credits, as the application process can be lengthy.

Pre-/Post-Series CNV Evaluation Project

This project is designed to evaluate the ability to effectively evaluate CNVs using the scoring metrics before and after a targeted educational intervention (the web series).  These results will also allow us to identify content areas that may require additional education or resources.

The option to participate in this project was made available to individuals registering for the CNV web series prior to December 19, 2019.  Assignments and instructions have been emailed to participants separately. Enrollment for this project is now CLOSED.

Please contact

Questions Related to Category 1: Initial Assessment of Genomic Content

Examples of functionally important elements include: 1) the SHOX enhancer recurrent 47.5 kb deletion that is approximately 160 kb downstream of the SHOX gene (Benito-Sanz et al., J Med Genet. 2012 Jul;49(7):442-50. PMID: 22791839) and 2) a duplication of the zone of polarizing activity regulatory sequence (ZRS) that affects the expression of the SHH gene

We encourage users to stay up-to-date on research involving variants in non-coding regions, positional effects, regulatory regions, etc., and to use clinical judgement when determining if the evidence is strong enough and/or the effects are well understood enough to warrant considering these things during clinical variant classification.  In this particular section, if one suspects the presence of a known functionally important element, the recommended next step is simply to continue with the rest of the evaluation (no points added); evidence supporting or refuting the role of the genomic elements contained within a particular CNV will be assessed in other aspects of the metric.

Questions Related to Category 2: Overlap with Established/Predicted Dosage Sensitive or Benign Genes/Genomic Regions

For the purposes of these scoring metrics, the term “established dosage sensitive” refers to those genes/genomic regions receiving a haploinsufficiency (HI) and/or triplosensitivity (TS) score of 3 by the ClinGen Dosage Sensitivity curation team.  A score of 3 indicates that there is sufficient evidence to support a dosage sensitivity mechanism for the gene/genomic region. “Established benign” refers to those genes/genomic regions receiving a HI and/or TS score of 40 (“dosage sensitivity unlikely”) by the ClinGen Dosage Sensitivity curation team.  ClinGen Dosage Sensitivity scores are publicly available here:

Prior to February 2019, ClinGen evaluated dosage sensitivity according to the procedures outlined in this publication:  Beginning in February 2019, ClinGen began evaluating single genes using an updated system that mirrors Category 4 of the scoring metrics.  Any single gene dosage sensitivity record dated February 2019 or later will have been evaluated using the same evidence evaluation framework as described in the scoring metrics.  An updated dosage sensitivity curation standard operating procedures document, outlining the updated processes for both single gene and recurrent region evaluation, is forthcoming.

No.  If a gene within your region already has a ClinGen HI Score of 3, do not also use category 2H because the gene is also “predicted” to be haploinsufficient.  This would be counting the same evidence concept twice.

No.  Category 2H points (0.15 total) can only be used once.

Questions Related to Category 3: Evaluation of Gene Number

When using Section 3 for CNV interpretation, if the CNV contains a large gene family of no or unknown clinical relevance (e.g., the olfactory receptor gene family), err on the side of caution and count that family as one gene. If the CNV contains a large gene family known to be associated with disease, such as the SCN gene cluster region at 2q24.3, each gene should be counted individually.

Questions Related to Category 4: Detailed Evaluation of Genomic Content Using Published Literature, Public Databases, and/or Internal Lab Data

To identify which gene may have the most supportive evidence available to initially take through the scoring metric, publicly available resources such as the ClinGen Dosage Map Resource, DECIPHER, and OMIM, are helpful.  Within DECIPHER, one may enter their coordinates of interest, then click on the “Genes” tab to see a list of all genes within the interval. This list can be filtered and sorted to quickly provide the user with information that may aid their prioritization.  For example, one can filter to see only those genes that are protein-coding, then sort the list to see those genes that have been designated “OMIM Morbid” at the top. If there are such genes in the interval, these may be a good place to start. Review of the information available in OMIM will provide information on whether or not the disease mechanism is consistent with loss of function or triplosensitivity; if disease mechanism is one of interest, evidence collection on the gene can be pursued further, if disease mechanism is not one of interest for the purposes of evaluating copy number variants (e.g., gain of function), this gene may be ruled out.

If there are no OMIM Morbid genes, next we suggest evaluating genes that are at least documented in OMIM; this information can be quickly gleaned from the DECIPHER gene table as well.  If this search fails to produce any genes with applicable human evidence, we would then suggest reviewing genes with pLI ≥ 0.90 (also available through the DECIPHER table). At each of these stages, we would search literature, public, and/or internal databases to determine if there was any human evidence available to support a relationship between haploinsufficiency/triplosensitivity and disease.

Potential differences between data obtained from the literature and data obtained from public databases lie in the amount of supporting data provided.  Complete information about the phenotype, the method of testing, previous testing, other variants identified, family history, evidence used to support the stated classification, etc. may be missing or absent in either source; however, these pieces of information in general tend to be more complete in case-reports from the literature.  Users should be aware of these potential limitations when scoring. For data available in public databases, consider the amount of supporting evidence available to allow for independent evaluation of the clinical classification provided.

As described in Supplement 1, Section 4, phenotypes that are “highly specific and relatively unique” are distinct, rarely seen in the general population, and typically observed in association with a single gene/condition; in some cases, these features may be pathognomonic.  Examples could include the lower lip pits associated with Van der Woude syndrome and variants in IRF6; facial trichilemmomas observed in Cowden syndrome and variants in PTEN, Kayser-Fleisher rings in Wilson’s disease and variants in ATP7B.  This could also include instances in which metabolic testing has already identified a specific deficits indicative of variation in a particular gene; for example, if a proband has demonstrated absence of alpha-galactosidase A via biochemical studies, that phenotype is highly specific and relatively unique to variants disrupting the corresponding GLA gene.

Phenotypes that are “highly specific, but not necessarily unique” are still relatively rare in the general population, but are not unique to any single genetic disorder. Examples of this may include early-onset epileptic encephalopathy, which has been associated with over 50 different genes.

Phenotypes that are not highly specific and/or are associated with high genetic heterogeneity include things that may be observed in the general population; may have significant; non-genetic influences; and may be a feature associated with many different genes and syndromes.  Phenotypes such as non-specific intellectual disability, autism, seizures (unspecified), which have been associated with hundreds of different genes, are examples of this category.

In this scenario, population data is being used to assert that, because a particular variant is common, it is unlikely to be associated with disease.  In general, even if a variant is very common in one particular ethnicity, but not observed in others, it is still too common to cause disease regardless of the ethnicity it was observed in, with few exceptions.  The ClinGen Sequence Variant Interpretation Working Group has published a recommendation that for BA1, the sequence guideline evidence category about population frequency that can result in a variant being automatically classified as benign, the frequency can be from any continental population with >2000 alleles tested, and that your patient’s ethnicity does not need to match that of this population in order to apply the rule (PMID: 30311383).  These recommendations can also be applied in this scenario.

Questions Related to Category 5: Evaluation of Inheritance Patterns/Family for Patient Being Studied

Since mosaic parents may be unaffected, partially affected, or completely affected, it is difficult to assign any points for evidence in this scenario.  However, points should not be deducted (in the context of “non-segregation”) if a parent with the CNV in the mosaic state does not exhibit the same clinical features as the proband with the same CNV observed constitutionally.  There are valid reasons why this would be the case (tissue distribution, level of mosaicism, etc.). The same logic applies for hemizygous CNVs inherited from unaffected heterozygous mothers - in general, these should not be considered “non-segregations,” and points should not be deducted.

Questions Related to Reporting

Uncoupling variant classification (Pathogenic through Benign) from clinical significance in the context of an individual patient’s diagnosis is key to objective and consistent interpretation of genomic variants.  This does not mean that case phenotype should not be taken into account when assessing evidence supporting the pathogenicity of a variant. It means that the classification of a variant should not be driven solely by the presentation of the patient under investigation (without consideration of other available evidence).  For example: There is compelling evidence in the literature that deletion of a particular gene results in Phenotype X; a laboratory evaluating a deletion of this gene is able to reach 0.99 points using the scoring metric, suggesting a classification of Pathogenic. The laboratory should not then classify the variant as “Uncertain Significance” (effectively discounting all previously collected evidence) because their patient did not display features of Phenotype X.  The presentation of their patient should be taken into account and scored as any other similar case would be scored, but it should not be the sole arbiter in the classification of this variant.

Other Questions

No.  ACMG and the Cancer Genomics Consortium (CGC) have put forth a separate technical standards document to address these variants: