H. Matsson, K. Tammimies, M. Zucchelli, H. Anthoni, P. Onkamo et al., , 2011.

, SNP variations in the 7q33 region containing DGKI are associated with dyslexia in the Finnish and German populations, Behav. Genet, vol.41, pp.134-140

W. Mclaren, L. Gil, S. E. Hunt, H. S. Riat, G. R. Ritchie et al., The Ensembl variant effect predictor, Genome Biol, vol.17, p.122, 2016.

R. L. Minster, N. L. Hawley, C. T. Su, G. Sun, E. E. Kershaw et al., A thrifty variant in CREBRF strongly influences body mass index in Samoans, Nat. Genet, vol.48, pp.1049-1054, 2016.

C. Moore, J. Sambrook, M. Walker, Z. Tolkien, S. Kaptoge et al., , 2014.

S. O'rahilly and I. S. Farooqi, Human obesity: a heritable neurobehavioral disorder that is highly sensitive to environmental conditions, Diabetes, vol.57, pp.2905-2910, 2008.

M. Pigeyre, F. T. Yazdi, Y. Kaur, and D. Meyre, Recent progress in genetics, epigenetics and metagenomics unveils the pathophysiology of human obesity, Clin. Sci. (Lond.), vol.130, pp.943-986, 2016.

A. Podcheko, P. Northcott, G. Bikopoulos, A. Lee, S. R. Bommareddi et al., , 2007.

A. L. Price, N. J. Patterson, R. M. Plenge, M. E. Weinblatt, N. A. Shadick et al., Principal components analysis corrects for stratification in genome-wide association studies, Nat. Genet, vol.38, pp.904-909, 2006.

S. Purcell, B. Neale, K. Todd-brown, L. Thomas, M. A. Ferreira et al., PLINK: a tool set for whole-genome association and population-based linkage analyses, Am. J. Hum. Genet, vol.81, pp.559-575, 2007.

S. M. Purcell, J. L. Moran, M. Fromer, D. Ruderfer, N. Solovieff et al., A polygenic burden of rare disruptive mutations in schizophrenia, Nature, vol.506, pp.185-190, 2014.

S. Saeed, A. Bonnefond, F. Tamanini, M. U. Mirza, J. Manzoor et al., Loss-offunction mutations in ADCY3 cause monogenic severe obesity, Nat. Genet, vol.50, pp.175-179, 2018.

K. E. Samocha, E. B. Robinson, S. J. Sanders, C. Stevens, A. Sabo et al.,

, A framework for the interpretation of de novo mutation in human disease, Nat. Genet, vol.46, pp.944-950

T. Singh, M. I. Kurki, D. Curtis, S. M. Purcell, L. Crooks et al.,

, INTERVAL Study; DDD Study

. Uk10-k-consortium, Rare loss-of-function variants in SETD1A are associated with schizophrenia and developmental disorders, Nat. Neurosci, vol.19, pp.571-577, 2016.

T. Singh, J. T. Walters, M. Johnstone, D. Curtis, J. Suvisaari et al., The contribution of rare variants to risk of schizophrenia in individuals with and without intellectual disability, Nat. Genet, vol.49, pp.1167-1173, 2017.

D. M. Styne, S. A. Arslanian, E. L. Connor, I. S. Farooqi, M. H. Murad et al., Pediatric obesity-assessment, treatment, and prevention: an endocrine society clinical practice guideline, J. Clin. Endocrinol. Metab, vol.102, pp.709-757, 2017.

V. Turcot, Y. Lu, H. M. Highland, C. Schurmann, A. E. Justice et al.,

. Epic-cvd and . Consortium, ExomeBP Consortium; Global Lipids Genetic Consortium; GoT2D Genes Consortium

, Protein-altering variants associated with body mass index implicate pathways that control energy intake and expenditure in obesity, Understanding Society Scientific Group, vol.50, pp.26-41, 2018.

G. A. Van-der-auwera, M. O. Carneiro, C. Hartl, R. Poplin, G. Del-angel et al., , 2013.

, From FastQ data to high confidence variant calls: the Genome Analysis Toolkit best practices pipeline, Curr. Protoc. Bioinformatics, vol.43, pp.119-132, 2015.

K. Walter, J. L. Min, J. Huang, L. Crooks, Y. Memari et al., The UK10K project identifies rare variants in health and disease, Nature, vol.526, pp.82-90, 2015.

C. Wang, X. Zhan, L. Liang, G. R. Abecasis, L. et al., Improved ancestry estimation for both genotyping and sequencing data using projection procrustes analysis and genotype imputation, Am. J. Hum. Genet, vol.96, pp.926-937, 2015.

E. Webster, M. T. Cho, N. Alexander, S. Desai, S. Naidu et al., De novo PHIP-predicted deleterious variants are associated with developmental delay, intellectual disability, obesity, and dysmorphic features, Cold Spring Harb. Mol. Case Stud, vol.2, p.1172, 2016.

E. Wheeler, N. Huang, E. G. Bochukova, J. M. Keogh, S. Lindsay et al., Genomewide SNP and CNV analysis identifies common and low-frequency variants associated with severe early-onset obesity, Nat. Genet, vol.45, pp.513-517, 2013.

J. K. White, A. K. Gerdin, N. A. Karp, E. Ryder, M. Buljan et al., Genome-wide generation and systematic phenotyping of knockout mice reveals new roles for many genes, Cell, vol.154, pp.452-464, 2013.

J. Yang, J. Seo, R. Nair, S. Han, S. Jang et al., DGKi regulates presynaptic release during mGluRdependent LTD, EMBO J, vol.30, pp.165-180, 2011.

, Thermo Fisher Scientific, 11668019) in serum-free Opti-MEM I medium (GIBCO, 31985) according to the manufacturer's protocols. After 6 h transfection media was replaced by full DMEM. The next day cells were incubated overnight with starvation media (DMEM no FBS) with or without the presence of 200ng/mL Leptin, Lipofectamine, 2000.

, Quantitation of firefly luciferase activity was performed using the Steadylite Plus Reporter Gene Assay System (Perkin Elmer, 6066759) according to the manufacturer's protocol

, 000 cells/well) and transiently transfected the next day with 50ng/well plasmid encoding empty pcDNA3.1(+) vector together with 50ng/well WT PHIP plasmid or 50ng/well WT PHIP plasmid together with different concentrations of mutant PHIP plasmid and combined with 50ng/well plasmid for Leptin Receptor, 50ng/well plasmid for POMC luciferase and 10ng/well plasmid for STAT3 using Lipofectamine 2000 in Opti-MEM I medium according to the manufacturer's protocols. After 6 h, transfection media was replaced by full DMEM. The next day cells were incubated overnight with starvation media (DMEM no FBS) with or without the presence of 200ng/mL Leptin. Quantitation of firefly luciferase activity was performed using the Steadylite Plus Reporter Gene Assay System, Dominant Negative Luciferase POMC Transcription Activation Assay HEK293 cells were seeded in white 96-well plates coated with Poly-D-Lysine, vol.40

, WT or mutant PHIP plasmid, combined with 50ng/well plasmid for Leptin Receptor, and 10ng/well plasmid for STAT3 using Lipofectamine 2000 in Opti-MEM I medium according to the manufacturer's protocols. After 6 h, transfection media was replaced by full DMEM. The next day cells were incubated overnight with starvation media and stimulated with 200ng/mL Leptin for 10 min. Cells were immediately fixed with 4% Formaldehyde in PBS for 20 min at room temperature, permeabilized with 0.2% Triton X-100 for 30 min at room temperature, blocked for 1 h in 3% BSA at room temperature and incubated overnight at 4 C with Mouse anti-HA tag (6E2) (Cell Signaling, 2367) in 1:100 dilution in 3% BSA. Cells were washed three times with PBS for 5 min, incubated with goat anti mouse secondary antibody Alexa Fluor 488 in 1:200 dilution in 3% BSA for 1 h at room temperature, washed 2 times with PBS for 5 min, incubated with DAPI in 1:500 dilution in PBS for 10 min and kept in PBS. Cells in the 96 well plates were imaged in the Opera Phenix High Content Screening Confocal system, obtaining 9 images per well. Quantification of nuclear and cytoplasmic localization was performed with the Harmony software (Perkin Elmer) using the Alexa 488 signal and nucleus to cytoplasm ratio was calculated by dividing the number of cells/well with positive signal in the nucleus (normalized to total number of cells in the well) to the number of cells/well with positive signal in the cytoplasm, Subcellular Localization of Human PHIP Variants COS-7 cells were seeded into black clear bottom CellCarrier-96 Ultra Microplates coated with Poly-D-Lysine (20.000 cells/well) or into glass coverslips in 12-well plates coated with Poly-D-Lysine (150.000 cells/well). Cells were transiently transfected with 100ng/ well plasmid encoding either empty pcDNA3.1(+) vector (negative control)

. Lee, After blocking with 5% milk solution in TBS-T for 1 h at room temperature, membranes were probed overnight at 4 C using Rabbit anti-PHIP (Abcam, ab86244) at 1:200 dilution in 5% milk in TBS-T. Cells were washed three times with TBS-T for 10 min at room temperature with gentle shaking and incubated with secondary antibody, Goat anti-rabbit IgG-HRP (Dako, P0448) diluted 1:2000 in 5% milk in TBS-T for 1 h at room temperature. Bands were developed using enhanced chemiluminescence (ECL) substrate (Promega, W1015) and images were captured with an ImageQuant LAS 4000 (GE Healthcare). The band intensity of western blots was quantified using FIJI. Leave-one-out Analyses To identify gene-based results driven by one or more variants, we applied the following leave-one-out strategy: 1-among the variants seen more than twice in our stage 1 sample (cases and controls together), we identified the variant with the lowest single-variant analysis p value whenever it is nominally significant (p < 0.05); 2-we removed this variant and performed the stage 1 gene-based test again. We repeated steps 1 and 2 until the stage 1 gene-based p value was above 0.1 or there were no additional variants seen more than twice and with a single-variant analysis p value < 0.05. For genes that were driven by one or two single variants (8 genes, Table S2; Figure S1), we genotyped single variants. Otherwise, we sequenced the coding region of the gene. Stage1+2 Meta-Analyses Gene-based meta-analysis was performed for 9 genes selected from stage 1 analysis and taken forward for targeted sequencing in stage 2. This analysis was performed using the MetaSKAT_wZ function from the R package MetaSKAT (version 0, Vitro Immunoprecipitation Assay HEK293 cells stably expressing the leptin receptor were seeded in 10cm cell culture dishes coated with Poly-D-Lysine (500.000 cells/ well). Cells were starved overnight, stimulated with 200ng/mL insulin (Sigma, i9278) or leptin for 15 min and lysed with cell lysis buffer containing 50 mM Tris, 50 mM KCL, 10 mM EDTA, 1% NP-40, vol.60, 2013.

, Genotyping Based on results from the stage 1 single-point analysis (47 variants, see below) and gene-based tests (14 variants, see below) (Figure S1), 53 variants were selected to take forward to stage 2 in an additional 1,810 SCOOP and 3,800 randomly-selected Fenland samples. Of the 53 variants, 48 assays were successfully designed for Agena genotyping

, Four SNPs failed QC resulting in 44 SNPs for single variant analysis. Sixty-two SCOOP cases and 23 Fenland controls with a call rate below 0.9 were removed, resulting in 1,754 SCOOP cases and 3,777 Fenland controls for single-variant analysis in the stage

, Genes known to harbor causal, highly penetrant mutations involved in human obesity were taken from Table 1 in Pigeyre, Tables S8A and S8B), 2016.

. Samocha, 2014) file: fordist_cleaned_nonpsych_z_pli_rec_null_data.txt. Two gene sets were created a constrained gene set where pLI > 09 (Tables S8B and S8D) and an unconstrained gene set with genes with pLI % 0.9, ) online system curates genes related to developmental delay and the strength of evidence for the association between the gene and developmental delay, vol.3, 2017.

. Purcell, We repeated primary analysis in patients with obesity and developmental delay (Table S9C) and with obesity alone (Table S9D). A secondary analysis was performed to assess genes, 2014.

. Briefly and . Seq, ) we calculated gene region test-statistics for an enrichment of genetic variants in cases compared to controls. For each gene, we evaluated the three analysis groupings used in the gene-based tests: BROAD (MAF < 1% and broadly damaging), STRICT (MAF < 0.025% and strictly damaging), and LOF (MAF < 0.025% and LoF)