Page 9 - Read Online

Aledo-Serrano et al. J Transl Genet Genom 2021;5:443-55 https://dx.doi.org/10.20517/jtgg.2021.40 Page 447

evidence suggests that there are few epilepsy-specific genes, and most of them are generally associated with
[34]
global neurodevelopmental diseases, causing both intellectual disability and autism spectrum disorders .
REVERSE PHENOTYPING
Reverse phenotyping refers to the refinement of phenotypes based on the interpretation of genetic data. The
strategy was firstly utilized in sarcoidosis patients and their phenotyping, and it is a promising new
[35]
approach in many other heterogeneous diseases, such as epilepsy or neurodevelopmental disorders
[Figure 1].

The picture of PCDH19-related epilepsy
The PCDH19 gene was initially included in the group of genes related to Dravet syndrome as it was
associated with early-onset febrile and prolonged seizures before two years of life and is usually associated
with cognitive and behavioral disturbances . Patients with pathogenic variants could be included into a
[36]
wide Dravet or “Dravet-like” phenotype. However, over time these patients were found to have very specific
characteristics. Most of the patients are girls, present seizures that occur in clusters, and have focal
semiology. In addition, cognitive and behavioral disorders are slightly different (including autistic spectrum
features without motor abnormalities), and the drugs that show efficacy are also different. Through a
reverse phenotyping process (from the gene to the syndrome and not from the syndrome to the gene, as
routinely performed), it was observed that there was a specific PCDH19-related epilepsy syndrome that
could be distinguished from the classical Dravet syndrome. Currently, it is known that PCDH19 is among
the 10 most frequent causes of genetic epilepsy with onset in the first two years of life causing a very
distinctive syndrome. This translates into different treatments . For example, levetiracetam is a drug with
[37]
low efficacy in Dravet syndrome and very high efficacy in PCDH19-related epilepsy .
[38]
Platforms to work with for reverse phenotyping
A major goal in the field of clinical genetics is to better understand the conundrums of genotype-phenotype
correlations. A detailed review of the recent literature and publicly available databases is necessary when
[39]
dealing with variants of uncertain significance . This process has been facilitated in recent years by the
emergence of bioinformatic tools, such as the R package VarformPDB, which organize genes and variants
associated to a condition, phenotype, or clinical feature from public databases such as OMIM (Online
Mendelian Inheritance in Man), Orphanet, ClinVar, Human Phenotype Ontology (HPO), UniProt
(Universal Protein Resource), and abstracts from PubMed . Moreover, the creation of huge biobanks
[40]
associating electronic medical record (EMR) data and genetics has led to improved clinical characterization
of variant carriers. In addition to deep clinical phenotyping, multi-omics strategies can be employed
through simultaneous compilation of biomarkers and samples from patients . Multiple innovative projects
[40]
have been made to ease EMR-based genomic research, such as the creation of the Precision Medicine
Initiative Cohort Programme, which allow both “direct genetic” and “reverse phenotyping”
implementations. The use of EHRs for genomic research presents multiple advantages, as large samples can
be recruited without the previous time-consuming and economic burden [40,41] .

The clinical genetics community employs numerous tools to address reverse phenotyping, including
DECIPHER, an open-access database that hosts a collection of tests for more than 37,000 patients with
[42]
genetic conditions to compare genotype-phenotype data . Additional examples, such as the Matchmaker
Exchange or GeneMatcher, serve as database mining using the association of genotype and phenotype
[43]
profiles . On the other hand, VarSome is a bioinformatic platform with tools aiming to compile
information needed for variant analysis . All these platforms allow the neurologist and geneticist to
[44]
perform direct and inverse phenotyping of patients, facilitating the interpretation of variants and the