CNVxplorer: a web tool to assist clinical interpretation of CNVs in rare disease patients

Requena, Francisco; Abdallah, Hamza Hadj; García, Alejandro Dorado; Nitschké, Patrick; Romana, Sergi; Malan, Valérie; Rausell, Antonio

doi:10.1101/2021.03.19.21253806

Cited by 3 publications

(7 citation statements)

References 73 publications

(79 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, it is important to assess the consistency of the patient's phenotype with the targeted genomic elements, as this can help prevent the misclassification as pathogenic of CNVs affecting Mendelian genes not associated with the clinical signs investigated. Indeed, this aspect has already been incorporated into current computational frameworks to assist the manual assessment of CNVs [3,13]; it would require the systematic incorporation of phenotypes into reference databases, such as Clinvar. Third, there is also a need for large-scale mutational scanning projects on cellular systems [14], to provide experimental evidence for CNVs not affecting coding genes.…”

Section: Discussionmentioning

confidence: 99%

“…To facilitate the use of CNVscore, we have implemented an application programming interface (API) supporting remote queries. In addition, CNVscore is integrated into the CNVxplorer framework for the clinical assessment of CNVs identified in patients with rare diseases (http://cnvxplorer.com, [3]). The complete list of CNVs identified in a patient's genome can, therefore, first be filtered by combining the pathogenicity and uncertainty CNVscores, to obtain a shortlist of highconfidence candidate pathogenic CNVs.…”

Section: Discussionmentioning

confidence: 99%

“…Various bioinformatics webtools [3,4] and pathogenicity scores (Table 1) have recently been developed to facilitate the annotation, filtering and clinical assessment of large numbers of candidate CNVs in patients with rare diseases. Two major categories of scores can be distinguished: (a) those implementing heuristics based on reference medical guidelines, such as ACMG and ClinGen [4] and (b) those based on supervised-learning approaches trained on reference pathogenic and non-pathogenic CNV datasets (Table 1).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

CNVscore calculates pathogenicity scores for copy number variants together with uncertainty estimates accounting for learning biases in reference Mendelian disorder datasets

Requena

Salgado

Malan

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Copy number variants (CNVs) are a major cause of rare pediatric diseases with a broad spectrum of phenotypes. Genetic diagnosis based on comparative genomic hybridization tests typically identifies ~8-10% of patients as having CNVs of unknown significance, revealing the current limits of clinical interpretation. The adoption of whole-genome sequencing (WGS) as a first-line genetic test has significantly increased the load of CNVs identified in single genomes. Alongside short- and long-read sequencing technologies, a number of pathogenicity scores have been developed for filtering and prioritizing large sets of candidate CNVs in clinical settings. However, current approaches are often based, either explicitly or implicitly, on clinically annotated reference sets, which are likely to bias their predictions. In this study we developed CNVscore, a supervised-learning approach combining tree ensembles and a Bayesian classifier trained on pathogenic and non-pathogenic CNVs from reference databases. Unlike previous approaches, CNVscore couples pathogenicity estimates with uncertainty scores, making it possible to evaluate the suitability of a model for the query CNVs. Comprehensive comparative benchmark tests across independent sets and against alternative methods showed that CNVscore effectively distinguishes between pathogenic and benign CNVs. We also found that CNVs associated with CNVscores of low uncertainty were predicted with significantly higher accuracy than those of high uncertainty. However, the performance of current scoring approaches, including CNVscore, was compromised on CNV sets enriched in highly uncertain variants and presenting unconventional features, such as functionally relevant non-coding elements or the presence of disease genes irrelevant for the clinical phenotypes investigated. Finally, we used the CNVscore framework to guide CNV scoring model selection for the French National Database of Constitutional CNVs (BANCCO), which includes clinical diagnosis annotations. The CNVscore framework provides an objective strategy for leveraging the uncertainty on bioinformatic predictions to enhance the assessment of CNV pathogenicity in rare-disease cohorts. CNVscore is available as open-source software from https://github.com/RausellLab/CNVscore and is integrated into the CNVxplorer webserver http://cnvxplorer.com.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

CNVscore calculates pathogenicity scores for copy number variants together with uncertainty estimates accounting for learning biases in reference Mendelian disorder datasets

Requena

Salgado

Malan

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…DisGeNET ( http://www.disgenet.org/web/DisGeNET/menu ) 88 , VarElect ( http://varelect.genecards.org/ ), and Schizophrenia Exome Sequencing Genebook 89 , 90 were also used to characterize variations. DECIPHER (DatabasE of genomiC varIation and Phenotype in Humans using Ensembl Resources; https://www.deciphergenomics.org ) 91 and CNVxplorer ( http://cnvxplorer.com ) 92 were used to study the pathogenicity and conservation of the identified CNVs. All genomic data for molecular variants in this study were compatible with Genome build GRCh37.…”

Section: Methodsmentioning

confidence: 99%

The conserved ASTN2/BRINP1 locus at 9q33.1–33.2 is associated with major psychiatric disorders in a large pedigree from Southern Spain

Pol-Fuster

Cañellas

Ruiz-Guerra

et al. 2021

Sci Rep

View full text Add to dashboard Cite

We investigated the genetic causes of major mental disorders (MMDs) including schizophrenia, bipolar disorder I, major depressive disorder and attention deficit hyperactive disorder, in a large family pedigree from Alpujarras, South of Spain, a region with high prevalence of psychotic disorders. We applied a systematic genomic approach based on karyotyping (n = 4), genotyping by genome-wide SNP array (n = 34) and whole-genome sequencing (n = 12). We performed genome-wide linkage analysis, family-based association analysis and polygenic risk score estimates. Significant linkage was obtained at chromosome 9 (9q33.1–33.2, LOD score = 4.11), a suggestive region that contains five candidate genes ASTN2, BRINP1, C5, TLR4 and TRIM32, previously associated with MMDs. Comprehensive analysis associated the MMD phenotype with genes of the immune system with dual brain functions. Moreover, the psychotic phenotype was enriched for genes involved in synapsis. These results should be considered once studying the genetics of psychiatric disorders in other families, especially the ones from the same region, since founder effects may be related to the high prevalence.

show abstract

“…One reason for this is the sheer volume of variational possibilities in genomic data. For example, it has been shown that CNVs are associated with several rare diseases (Li et al ., 2020), but such analyses either use simple statistics or require visual inspection of CNV distributions (Requena et al ., 2021; Collins et al ., 2020; MacDonald et al ., 2014). Since there are many types of CNVs and SVs (deletions, inversions, tandem duplications, translocations), manual development of machine learning features requires time-consuming, error-prone, experts’ labor.…”

Section: Introductionmentioning

confidence: 99%

DBFE: Distribution-based feature extraction from copy number and structural variants in whole-genome data

Piernik

Brzeziński

Sztromwasser³

et al. 2022

Preprint

View full text Add to dashboard Cite

Motivation: Whole-genome sequencing has revolutionized biosciences by providing tools for constructing complete DNA sequences of individuals. With entire genomes at hand, scientists can pinpoint DNA fragments responsible for different cancers and predict patient responses to cancer treatments. However, the sheer volume of whole-genome data makes it difficult to encode the characteristics of genomic variants as features for machine learning algorithms. Results: We present three feature extraction methods that facilitate classifier learning from distributions of genomic variants. The proposed approaches use binning, clustering, and kernel density estimation to produce features that discriminate between two groups of patients. Experiments on genomes of 219 ovarian, 61 lung, and 929 breast cancer patients show that the proposed approaches automatically identify genomic biomarkers associated with cancer subtypes and clinical response to oncological treatment. Finally, we show that the extracted features can be used alongside unsupervised learning methods to analyze genomic samples

show abstract

CNVxplorer: a web tool to assist clinical interpretation of CNVs in rare disease patients

Cited by 3 publications

References 73 publications

CNVscore calculates pathogenicity scores for copy number variants together with uncertainty estimates accounting for learning biases in reference Mendelian disorder datasets

CNVscore calculates pathogenicity scores for copy number variants together with uncertainty estimates accounting for learning biases in reference Mendelian disorder datasets

The conserved ASTN2/BRINP1 locus at 9q33.1–33.2 is associated with major psychiatric disorders in a large pedigree from Southern Spain

DBFE: Distribution-based feature extraction from copy number and structural variants in whole-genome data

Contact Info

Product

Resources

About