Syndromic obesity are rare forms of obesity (1% of cases), involving severe obesity and early to multi organ involvement (mental retardation, dysmorphic, sensorineural damage and / or endocrine). To date, the genetic defects are identified in only 5% of cases (Prader-Willi syndrome, Bardet-Biedl syndrome, mutation of leptin or its receptor, the proconvertase-1, proopiomelanocortin or SIM-1 and TRKB genes, high resolution karyotype or abnormal DNA chips, ...). Precocity and severity of obesity are those for a little dependent genetic environment.

The investigators aim is to identify new gene variants in subjects with syndromic obesity sharing common phenotypic features.

Sequencing of exons :

To date, most of syndromic obesities have not been elucidated. But, the lack of identification of molecular abnormalities involved limits information to families about the risk of recurrence and optimal support of patient.

A new molecular biology tool called sequencing of exons ("exome") or targeted sequencing subset of the genome that code for proteins showed its power to identify mutations responsible for rare diseases, in a small number of individuals achieved unrelated. Indeed, this tool helped to identify two mutations in the gene TSPAN12 (Tetraspanin 12) in 23 subjects with exudative vitreo-retinopathy, 3 mutations in the gene PRRT2 (proline-rich transmembrane protein 2) in 8 Chinese families with kinesigenic paroxysmal dyskinesia, 2 nonsense mutations in the gene ANGPTL3 (angiopoietin-like 3) in 2 subjects with familial combined hypolipidemia, a mutation of the gene RBM10 (RNA binding motif 10) in 3 patients with syndromic form of cleft palate (TARP syndrome) or a mutation of the gene PALB2 (partner and localizer of BRCA2) in a patient with pancreatic cancer. The investigators believe that this tool could identify new molecular abnormalities in individuals with syndromic obesity. The detailed analysis of phenotypic data could identify a group of obese patients (at least 10) sharing clinical and biological phenotype and in which this technique would be used.

The exome sequencing will be conducted in subjects selected from their phenotype but also among their relatives (2 parents and a sibling) to provide additional discriminative capacity. Every new variant found at the homozygous state in obese patients and unaffected relatives will be immediately excluded from the list of candidate mutations. Conversely, any variant found in patients but absent in both parents and sibling is a potential de novo mutation. The presence of any variant candidate will be sought in the different exome public databases (dbSNP, 1000 Genomes, Exome Variant Server) and, in the case of an unlisted variant, it will be possible to genotype in cohorts of children with common obesity and non-obese in order to determine its frequency and thus provide additional information about its potential involvement in the syndromic obesity phenotype. The last step that will validate the role of the variant gene in the clinical picture will be in vitro analysis of the functional consequences on the protein and then in vivo in an animal models.

The sequencing of all exons ("exomes") of the 40 selected subjects (10 probands with both parents and non-obese sibling) will be performed on genomic platform using a new type of generation sequencer Illumina HiSeq 2000. To do this, DNA 3μg (15μL concentration to 200 ng/uL) will be needed per sample with then sequencing of type "paired-end" on 2 x 75bp. The generated sequences will be aligned with the reference genome (build hg19) using the BWA Software. Duplicates (PCR duplicates' and 'optical duplicates') will be eliminated as non-paired and misaligned sequences (Q-score <20). The generated sequences will then be recalibrated by GATK software to improve detection of such variations insertion / deletion. Detecting variants will be carried out using tools software and annotation will use the annovar software. Analyzes will focus on detected variants with a depth of at least 10x. The investigators will select any variant found in the homozygous state in the patient and in the heterozygous state with his parents (recessive hypothesis). Variants also found homozygous among the unaffected relatives will be excluded from the list of candidates. Any variant found in patients but absent in both parents may also be a potential candidate in case de novo The investigators believe that this strategy could facilitate diagnosis but also, after validation of the role of this variant in the pathology, assess the risk of recurrence for families by offering prenatal diagnosis if necessary. It also will provide a support as early as possible to these children, like the medical, psychological and social care of children with Prader-Willi syndrome. Indeed, the existence of a genetic diagnosis responsible for the phenotype allows the development of specific life project adapted to the patient and facilitates the development aid for disability through the reference centers and collaboration with departmental homes of disabled people (MDPH). Finally, the progress in understanding the mechanisms of these obesities may help to better understand the pathophysiology of the most common forms of obesity and improve the management.

Patient Selection :

For children and adults with syndromic obesity, day hospitalization is systematically planned in usual care to perform a complete clinical and biological assessment. The information about the various examinations made during the day of hospitalization are given during a consultation. The genetic study will be proposed It was proposed during this consultation to patients, children and adults, and their families. The hospitalization date is usually in the month following the consultation.

Inclusion - patient day hospitalization :

As part of usual care, a precise phenotypic characterization is performed by the investigator day hospitalization : family weight history, retrospective analysis of the personal weight history, clinical examination, anthropometry, dietary consultation, measures of body composition and expenditure energy of rest, food and physical activity questionnaires, endocrine and metabolic balanced.

The research will consist of a blood sample of 20 mL extra for the genetic study and will be conducted in obese subjects after signing a consent. Patients for whom analyzable samples are already stored in the DNA bank No. DC2009-957 will not be sampled again.

Inclusion Criteria:

• Patient age ≥ 1 year with no upper age limit, with early onset (age of obesity onset <6 years) and severe (Z-score of BMI ≥ + 3DS) obesity associated with psychomotor retardation or mental , food impulsiveness, +/- dysmorphy and / or endocrine abnormalities

Exclusion Criteria:

• Identified syndromic obesity

10 unrelated subjects sharing phenotypic features will be selected as their two parents and one unaffected sibling corresponding to a total of 40 subjects.