comorbidPGS: an R package assessing shared predisposition between Phenotypes using Polygenic Scores.

Introduction Polygenic Score (PGS) is a valuable method for assessing the estimated genetic liability to a given outcome or genetic variability contributing to a quantitative trait. While PRSs are widely used for complex traits, their application in uncovering shared genetic predisposition between phenotypes, i.e. when genetic variants influence more than one phenotype, remains limited. Methods We developed an R package, comorbidPGS, which facilitates a systematic evaluation of shared genetic effects among (cor)related phenotypes using PGSs. The comorbidPGS package takes as input a set of Single Nucleotide Polymorphisms (SNPs) along with their established effects on the original phenotype (Po), referred to as Po-PGS. It generates a comprehensive summary of effect(s) of Po-PGS on target phenotype(s) (Pt) with customisable graphical features. Results We applied comorbidPGS to investigate the shared genetic predisposition between phenotypes defining elevated blood pressure (Systolic Blood Pressure, SBP; Diastolic Blood Pressure, DBP; Pulse Pressure, PP) and several cancers (Breast Cancer, BrC; Pancreatic Cancer, PanC; Kidney Cancer, KidC; Prostate Cancer, PrC; Colorectal Cancer, CrC) using the European ancestry UK Biobank individuals and GWAS meta-analyses summary statistics from independent set of European ancestry individuals. We report a significant association between elevated DBP and the genetic risk of PrC (ß (SE)=0.066 (0.017), P-value=9.64×10^(-5)), as well as between CrC PGS and both, lower SBP (ß (SE)=-0.10 [0.029], P-value=3.83×10^(-4))) and lower DBP (ß (SE)=-0.055 [0.017], P-value=1.05×10^(-3)). Our analysis highlights two nominally significant relationships for individuals with genetic predisposition to elevated SBP leading to higher risk of KidC (OR [95%CI]=1.04 [1.0039-1.087], P-value=2.82×10^(-2)) and PrC (OR [95%CI]=1.02 [1.003-1.041], P-value=2.22×10^(-2)). Conclusion Using comorbidPGS, we underscore mechanistic relationships between blood pressure regulation and susceptibility to three comorbid malignancies. This package offers valuable means to evaluate shared genetic susceptibility between (cor)related phenotypes through polygenic scores.

A framework for conducting time-varying genome-wide association studies: An application to body mass index across childhood in six multiethnic cohorts.

Genetic effects on changes in human traits over time are understudied and may have important pathophysiological impact. We propose a framework that enables data quality control, implements mixed models to evaluate trajectories of change in traits, and estimates phenotypes to identify age-varying genetic effects in genome-wide association studies (GWASs). Using childhood body mass index (BMI) as an example, we included 71,336 participants from six cohorts and estimated the slope and area under the BMI curve within four time periods (infancy, early childhood, late childhood and adolescence) for each participant, in addition to the age and BMI at the adiposity peak and the adiposity rebound. GWAS on each of the estimated phenotypes identified 28 genome-wide significant variants at 13 loci across the 12 estimated phenotypes, one of which was novel (in DAOA) and had not been previously associated with childhood or adult BMI. Genetic studies of changes in human traits over time could uncover novel biological mechanisms influencing quantitative traits.

Trans-ancestral genome-wide association study of longitudinal pubertal height growth and shared heritability with adult health outcomes.

BACKGROUND: Pubertal growth patterns correlate with future health outcomes. However, the genetic mechanisms mediating growth trajectories remain largely unknown. Here, we modeled longitudinal height growth with Super-Imposition by Translation And Rotation (SITAR) growth curve analysis on ~ 56,000 trans-ancestry samples with repeated height measurements from age 5 years to adulthood. We performed genetic analysis on six phenotypes representing the magnitude, timing, and intensity of the pubertal growth spurt. To investigate the lifelong impact of genetic variants associated with pubertal growth trajectories, we performed genetic correlation analyses and phenome-wide association studies in the Penn Medicine BioBank and the UK Biobank. RESULTS: Large-scale growth modeling enables an unprecedented view of adolescent growth across contemporary and 20th-century pediatric cohorts. We identify 26 genome-wide significant loci and leverage trans-ancestry data to perform fine-mapping. Our data reveals genetic relationships between pediatric height growth and health across the life course, with different growth trajectories correlated with different outcomes. For instance, a faster tempo of pubertal growth correlates with higher bone mineral density, HOMA-IR, fasting insulin, type 2 diabetes, and lung cancer, whereas being taller at early puberty, taller across puberty, and having quicker pubertal growth were associated with higher risk for atrial fibrillation. CONCLUSION: We report novel genetic associations with the tempo of pubertal growth and find that genetic determinants of growth are correlated with reproductive, glycemic, respiratory, and cardiac traits in adulthood. These results aid in identifying specific growth trajectories impacting lifelong health and show that there may not be a single "optimal" pubertal growth pattern.

51st European Mathematical Genetics Meeting (EMGM) 2023.

NA.

Bidirectional Mendelian Randomization and Multiphenotype GWAS Show Causality and Shared Pathophysiology Between Depression and Type 2 Diabetes.

OBJECTIVE: Depression is a common comorbidity of type 2 diabetes. We assessed the causal relationships and shared genetics between them. RESEARCH DESIGN AND METHODS: We applied two-sample, bidirectional Mendelian randomization (MR) to assess causality between type 2 diabetes and depression. We investigated potential mediation using two-step MR. To identify shared genetics, we performed 1) genome-wide association studies (GWAS) separately and 2) multiphenotype GWAS (MP-GWAS) of type 2 diabetes (19,344 case subjects, 463,641 control subjects) and depression using major depressive disorder (MDD) (5,262 case subjects, 86,275 control subjects) and self-reported depressive symptoms (n = 153,079) in the UK Biobank. We analyzed expression quantitative trait locus (eQTL) data from public databases to identify target genes in relevant tissues. RESULTS: MR demonstrated a significant causal effect of depression on type 2 diabetes (odds ratio 1.26 [95% CI 1.11-1.44], P = 5.46 × 10-4) but not in the reverse direction. Mediation analysis indicated that 36.5% (12.4-57.6%, P = 0.0499) of the effect from depression on type 2 diabetes was mediated by BMI. GWAS of type 2 diabetes and depressive symptoms did not identify shared loci. MP-GWAS identified seven shared loci mapped to TCF7L2, CDKAL1, IGF2BP2, SPRY2, CCND2-AS1, IRS1, CDKN2B-AS1. MDD has not brought any significant association in either GWAS or MP-GWAS. Most MP-GWAS loci had an eQTL, including single nucleotide polymorphisms implicating the cell cycle gene CCND2 in pancreatic islets and brain and the insulin signaling gene IRS1 in adipose tissue, suggesting a multitissue and pleiotropic underlying mechanism. CONCLUSIONS: Our results highlight the importance to prevent type 2 diabetes at the onset of depressive symptoms and the need to maintain a healthy weight in the context of its effect on depression and type 2 diabetes comorbidity.

Abdominal obesity is a more important causal risk factor for pancreatic cancer than overall obesity.

Obesity and type 2 diabetes (T2D) are associated with increased risk of pancreatic cancer. Here we assessed the relationship between pancreatic cancer and two distinct measures of obesity, namely total adiposity, using BMI, versus abdominal adiposity, using BMI adjusted waist-to-hip ratio (WHRadjBMI) by utilising polygenic scores (PGS) and Mendelian randomisation (MR) analyses. We constructed z-score weighted PGS for BMI and WHRadjBMI using publicly available data and tested for their association with pancreatic cancer defined in UK biobank (UKBB). Using publicly available summary statistics, we then performed bi-directional MR analyses between the two obesity traits and pancreatic cancer. PGSBMI was significantly (multiple testing-corrected) associated with pancreatic cancer (OR[95%CI] = 1.0804[1.025-1.14], P = 0.0037). The significance of association declined after T2D adjustment (OR[95%CI] = 1.073[1.018-1.13], P = 0.00904). PGSWHRadjBMI association with pancreatic cancer was at the margin of statistical significance (OR[95%CI] = 1.047[0.99-1.104], P = 0.086). T2D adjustment effectively lost any suggestive association of PGSWHRadjBMI with pancreatic cancer (OR[95%CI] = 1.039[0.99-1.097], P = 0.14). MR analyses showed a nominally significant causal effect of WHRadjBMI on pancreatic cancer (OR[95%CI] = 1.00095[1.00011-1.0018], P = 0.027) but not for BMI on pancreatic cancer. Overall, we show that abdominal adiposity measured using WHRadjBMI, may be a more important causal risk factor for pancreatic cancer compared to total adiposity, with T2D being a potential driver of this relationship.

Genome-wide association analyses of physical activity and sedentary behavior provide insights into underlying mechanisms and roles in disease prevention.

Although physical activity and sedentary behavior are moderately heritable, little is known about the mechanisms that influence these traits. Combining data for up to 703,901 individuals from 51 studies in a multi-ancestry meta-analysis of genome-wide association studies yields 99 loci that associate with self-reported moderate-to-vigorous intensity physical activity during leisure time (MVPA), leisure screen time (LST) and/or sedentary behavior at work. Loci associated with LST are enriched for genes whose expression in skeletal muscle is altered by resistance training. A missense variant in ACTN3 makes the alpha-actinin-3 filaments more flexible, resulting in lower maximal force in isolated type IIA muscle fibers, and possibly protection from exercise-induced muscle damage. Finally, Mendelian randomization analyses show that beneficial effects of lower LST and higher MVPA on several risk factors and diseases are mediated or confounded by body mass index (BMI). Our results provide insights into physical activity mechanisms and its role in disease prevention.

LongITools: Dynamic longitudinal exposome trajectories in cardiovascular and metabolic noncommunicable diseases.

The current epidemics of cardiovascular and metabolic noncommunicable diseases have emerged alongside dramatic modifications in lifestyle and living environments. These correspond to changes in our "modern" postwar societies globally characterized by rural-to-urban migration, modernization of agricultural practices, and transportation, climate change, and aging. Evidence suggests that these changes are related to each other, although the social and biological mechanisms as well as their interactions have yet to be uncovered. LongITools, as one of the 9 projects included in the European Human Exposome Network, will tackle this environmental health equation linking multidimensional environmental exposures to the occurrence of cardiovascular and metabolic noncommunicable diseases.

Bifidobacterium Is Enriched in Gut Microbiome of Kashmiri Women with Polycystic Ovary Syndrome.

Polycystic ovary syndrome (PCOS) is a very common endocrine condition in women in India. Gut microbiome alterations were shown to be involved in PCOS, yet it is remarkably understudied in Indian women who have a higher incidence of PCOS as compared to other ethnic populations. During the regional PCOS screening program among young women, we recruited 19 drug naive women with PCOS and 20 control women at the Sher-i-Kashmir Institute of Medical Sciences, Kashmir, North India. We profiled the gut microbiome in faecal samples by 16S rRNA sequencing and included 40/58 operational taxonomic units (OTUs) detected in at least 1/3 of the subjects with relative abundance (RA) ≥ 0.1%. We compared the RAs at a family/genus level in PCOS/non-PCOS groups and their correlation with 33 metabolic and hormonal factors, and corrected for multiple testing, while taking the variation in day of menstrual cycle at sample collection, age and BMI into account. Five genera were significantly enriched in PCOS cases: Sarcina, Megasphaera, and previously reported for PCOS Bifidobacterium, Collinsella and Paraprevotella confirmed by different statistical models. At the family level, the relative abundance of Bifidobacteriaceae was enriched, whereas Peptococcaceae was decreased among cases. We observed increased relative abundance of Collinsella and Paraprevotella with higher fasting blood glucose levels, and Paraprevotella and Alkalibacterium with larger hip, waist circumference, weight, and Peptococcaceae with lower prolactin levels. We also detected a novel association between Eubacterium and follicle-stimulating hormone levels and between Bifidobacterium and alkaline phosphatase, independently of the BMI of the participants. Our report supports that there is a relationship between gut microbiome composition and PCOS with links to specific reproductive health metabolic and hormonal predictors in Indian women.

Relationship between glucose homeostasis and obesity in early life-a study of Italian children and adolescents.

Epidemic obesity is the most important risk factor for prediabetes and type 2 diabetes (T2D) in youth as it is in adults. Obesity shares pathophysiological mechanisms with T2D and is likely to share part of the genetic background. We aimed to test if weighted genetic risk scores (GRSs) for T2D, fasting glucose (FG) and fasting insulin (FI) predict glycaemic traits and if there is a causal relationship between obesity and impaired glucose metabolism in children and adolescents. Genotyping of 42 SNPs established by genome-wide association studies for T2D, FG and FI was performed in 1660 Italian youths aged between 2 and 19 years. We defined GRS for T2D, FG and FI and tested their effects on glycaemic traits, including FG, FI, indices of insulin resistance/beta cell function and body mass index (BMI). We evaluated causal relationships between obesity and FG/FI using one-sample Mendelian randomization analyses in both directions. GRS-FG was associated with FG (beta = 0.075 mmol/l, SE = 0.011, P = 1.58 × 10-11) and beta cell function (beta = -0.041, SE = 0.0090 P = 5.13 × 10-6). GRS-T2D also demonstrated an association with beta cell function (beta = -0.020, SE = 0.021 P = 0.030). We detected a causal effect of increased BMI on levels of FI in Italian youths (beta = 0.31 ln (pmol/l), 95%CI [0.078, 0.54], P = 0.0085), while there was no effect of FG/FI levels on BMI. Our results demonstrate that the glycaemic and T2D risk genetic variants contribute to higher FG and FI levels and decreased beta cell function in children and adolescents. The causal effects of adiposity on increased insulin resistance are detectable from childhood age.

The trans-ancestral genomic architecture of glycemic traits.

Glycemic traits are used to diagnose and monitor type 2 diabetes and cardiometabolic health. To date, most genetic studies of glycemic traits have focused on individuals of European ancestry. Here we aggregated genome-wide association studies comprising up to 281,416 individuals without diabetes (30% non-European ancestry) for whom fasting glucose, 2-h glucose after an oral glucose challenge, glycated hemoglobin and fasting insulin data were available. Trans-ancestry and single-ancestry meta-analyses identified 242 loci (99 novel; P < 5 × 10-8), 80% of which had no significant evidence of between-ancestry heterogeneity. Analyses restricted to individuals of European ancestry with equivalent sample size would have led to 24 fewer new loci. Compared with single-ancestry analyses, equivalent-sized trans-ancestry fine-mapping reduced the number of estimated variants in 99% credible sets by a median of 37.5%. Genomic-feature, gene-expression and gene-set analyses revealed distinct biological signatures for each trait, highlighting different underlying biological pathways. Our results increase our understanding of diabetes pathophysiology by using trans-ancestry studies for improved power and resolution.

Risk of Recurrent Pregnancy Loss in the Ukrainian Population Using a Combined Effect of Genetic Variants: A Case-Control Study.

We assessed the predictive ability of a combined genetic variant panel for the risk of recurrent pregnancy loss (RPL) through a case-control study. Our study sample was from Ukraine and included 114 cases with idiopathic RPL and 106 controls without any pregnancy losses/complications and with at least one healthy child. We genotyped variants within 12 genetic loci reflecting the main biological pathways involved in pregnancy maintenance: blood coagulation (F2, F5, F7, GP1A), hormonal regulation (ESR1, ADRB2), endometrium and placental function (ENOS, ACE), folate metabolism (MTHFR) and inflammatory response (IL6, IL8, IL10). We showed that a genetic risk score (GRS) calculated from the 12 variants was associated with an increased risk of RPL (odds ratio 1.56, 95% CI: 1.21, 2.04, p = 8.7 × 10-4). The receiver operator characteristic (ROC) analysis resulted in an area under the curve (AUC) of 0.64 (95% CI: 0.57, 0.72), indicating an improved ability of the GRS to classify women with and without RPL. Ιmplementation of the GRS approach can help define women at higher risk of complex multifactorial conditions such as RPL. Future well-powered genome-wide association studies will help in dissecting biological pathways previously unknown for RPL and further improve the identification of women with RPL susceptibility.

Alzheimer's disease pathology explains association between dementia with Lewy bodies and APOE-ε4/TOMM40 long poly-T repeat allele variants.

INTRODUCTION: The role of TOMM40-APOE 19q13.3 region variants is well documented in Alzheimer's disease (AD) but remains contentious in dementia with Lewy bodies (DLB) and Parkinson's disease dementia (PDD). METHODS: We dissected genetic profiles within the TOMM40-APOE region in 451 individuals from four European brain banks, including DLB and PDD cases with/without neuropathological evidence of AD-related pathology and healthy controls. RESULTS: TOMM 40-L/APOE-ε4 alleles were associated with DLB (OR TOMM40 -L = 3.61; P value = 3.23 × 10-9; OR APOE -ε4 = 3.75; P value = 4.90 × 10-10) and earlier age at onset of DLB (HR TOMM40 -L = 1.33, P value = .031; HR APOE -ε4 = 1.46, P value = .004), but not with PDD. The TOMM40-L/APOE-ε4 effect was most pronounced in DLB individuals with concomitant AD pathology (OR TOMM40 -L = 4.40, P value = 1.15 × 10-6; OR APOE - ε 4 = 5.65, P value = 2.97 × 10-8) but was not significant in DLB without AD. Meta-analyses combining all APOE-ε4 data in DLB confirmed our findings (ORDLB = 2.93, P value = 3.78 × 10-99; ORDLB+AD = 5.36, P value = 1.56 × 10-47). DISCUSSION: APOE-ε4/TOMM 40-L alleles increase susceptibility and risk of earlier DLB onset, an effect explained by concomitant AD-related pathology. These findings have important implications in future drug discovery and development efforts in DLB.

GWAS on longitudinal growth traits reveals different genetic factors influencing infant, child, and adult BMI.

Early childhood growth patterns are associated with adult health, yet the genetic factors and the developmental stages involved are not fully understood. Here, we combine genome-wide association studies with modeling of longitudinal growth traits to study the genetics of infant and child growth, followed by functional, pathway, genetic correlation, risk score, and colocalization analyses to determine how developmental timings, molecular pathways, and genetic determinants of these traits overlap with those of adult health. We found a robust overlap between the genetics of child and adult body mass index (BMI), with variants associated with adult BMI acting as early as 4 to 6 years old. However, we demonstrated a completely distinct genetic makeup for peak BMI during infancy, influenced by variation at the LEPR/LEPROT locus. These findings suggest that different genetic factors control infant and child BMI. In light of the obesity epidemic, these findings are important to inform the timing and targets of prevention strategies.

Genetic determinants of risk in pulmonary arterial hypertension: international genome-wide association studies and meta-analysis.

BACKGROUND: Rare genetic variants cause pulmonary arterial hypertension, but the contribution of common genetic variation to disease risk and natural history is poorly characterised. We tested for genome-wide association for pulmonary arterial hypertension in large international cohorts and assessed the contribution of associated regions to outcomes. METHODS: We did two separate genome-wide association studies (GWAS) and a meta-analysis of pulmonary arterial hypertension. These GWAS used data from four international case-control studies across 11 744 individuals with European ancestry (including 2085 patients). One GWAS used genotypes from 5895 whole-genome sequences and the other GWAS used genotyping array data from an additional 5849 individuals. Cross-validation of loci reaching genome-wide significance was sought by meta-analysis. Conditional analysis corrected for the most significant variants at each locus was used to resolve signals for multiple associations. We functionally annotated associated variants and tested associations with duration of survival. All-cause mortality was the primary endpoint in survival analyses. FINDINGS: A locus near SOX17 (rs10103692, odds ratio 1·80 [95% CI 1·55-2·08], p=5·13 × 10-15) and a second locus in HLA-DPA1 and HLA-DPB1 (collectively referred to as HLA-DPA1/DPB1 here; rs2856830, 1·56 [1·42-1·71], p=7·65 × 10-20) within the class II MHC region were associated with pulmonary arterial hypertension. The SOX17 locus had two independent signals associated with pulmonary arterial hypertension (rs13266183, 1·36 [1·25-1·48], p=1·69 × 10-12; and rs10103692). Functional and epigenomic data indicate that the risk variants near SOX17 alter gene regulation via an enhancer active in endothelial cells. Pulmonary arterial hypertension risk variants determined haplotype-specific enhancer activity, and CRISPR-mediated inhibition of the enhancer reduced SOX17 expression. The HLA-DPA1/DPB1 rs2856830 genotype was strongly associated with survival. Median survival from diagnosis in patients with pulmonary arterial hypertension with the C/C homozygous genotype was double (13·50 years [95% CI 12·07 to >13·50]) that of those with the T/T genotype (6·97 years [6·02-8·05]), despite similar baseline disease severity. INTERPRETATION: This is the first study to report that common genetic variation at loci in an enhancer near SOX17 and in HLA-DPA1/DPB1 is associated with pulmonary arterial hypertension. Impairment of SOX17 function might be more common in pulmonary arterial hypertension than suggested by rare mutations in SOX17. Further studies are needed to confirm the association between HLA typing or rs2856830 genotyping and survival, and to determine whether HLA typing or rs2856830 genotyping improves risk stratification in clinical practice or trials. FUNDING: UK NIHR, BHF, UK MRC, Dinosaur Trust, NIH/NHLBI, ERS, EMBO, Wellcome Trust, EU, AHA, ACClinPharm, Netherlands CVRI, Dutch Heart Foundation, Dutch Federation of UMC, Netherlands OHRD and RNAS, German DFG, German BMBF, APH Paris, INSERM, Université Paris-Sud, and French ANR.

A meta-analysis of Italian and Estonian individuals shows an effect of common variants in HMGCR on blood apoB levels.

Aim: The aim of the study was to explore the effects of variants at HMGCR-KIF6loci on a range of cardio-metabolic phenotypes. Methods: We analyzed the range of variants within Genetics in Brisighella Health Study and KIF6 genes using an additive genetic model on 18 cardiometabolic phenotypes in a sample of 1645 individuals from the Genetics in Brisighella Health Study and replicated in 10,662 individuals from the Estonian Genome Center University of Tartu. Results: We defined directly the effects of rs3846662:C>A at HMGCR on apoB levels. The analysis also confirmed effects of on low-density lipoprotein-cholesterol and total cholesterol levels. Variants in KIF6 gene did not reveal any associations with cardiometabolic phenotypes. Conclusion: This study highlights effect of HMGCR locus on assay-determined apoB levels, an infrequent measure of blood lipids in large studies.

A neurobiological pathway to smoking in adolescence: TTC12-ANKK1-DRD2 variants and reward response.

The TTC12-ANKK1-DRD2 gene-cluster has been implicated in adult smoking. Here, we investigated the contribution of individual genes in the TTC12-ANKK1-DRD2 cluster in smoking and their association with smoking-associated reward processing in adolescence. A meta-analysis of TTC12-ANKK1-DRD2 variants and self-reported smoking behaviours was performed in four European adolescent cohorts (N = 14,084). The minor G-allele of rs2236709, mapping TTC12, was associated with self-reported smoking (p = 5.0 × 10-4) and higher plasma cotinine levels (p = 7.0 × 10-5). This risk allele was linked to an increased ventral-striatal blood-oxygen level-dependent (BOLD) response during reward anticipation (n = 1,263) and with higher DRD2 gene expression in the striatum (p = 0.013), but not with TTC12 or ANKK gene expression. These data suggest a role for the TTC12-ANKK1-DRD2 gene-cluster in adolescent smoking behaviours, provide evidence for the involvement of DRD2 in the early stages of addiction and support the notion that genetically-driven inter-individual differences in dopaminergic transmission mediate reward sensitivity and risk to smoking.

An Expanded Genome-Wide Association Study of Type 2 Diabetes in Europeans.

To characterize type 2 diabetes (T2D)-associated variation across the allele frequency spectrum, we conducted a meta-analysis of genome-wide association data from 26,676 T2D case and 132,532 control subjects of European ancestry after imputation using the 1000 Genomes multiethnic reference panel. Promising association signals were followed up in additional data sets (of 14,545 or 7,397 T2D case and 38,994 or 71,604 control subjects). We identified 13 novel T2D-associated loci (P < 5 × 10-8), including variants near the GLP2R, GIP, and HLA-DQA1 genes. Our analysis brought the total number of independent T2D associations to 128 distinct signals at 113 loci. Despite substantially increased sample size and more complete coverage of low-frequency variation, all novel associations were driven by common single nucleotide variants. Credible sets of potentially causal variants were generally larger than those based on imputation with earlier reference panels, consistent with resolution of causal signals to common risk haplotypes. Stratification of T2D-associated loci based on T2D-related quantitative trait associations revealed tissue-specific enrichment of regulatory annotations in pancreatic islet enhancers for loci influencing insulin secretion and in adipocytes, monocytes, and hepatocytes for insulin action-associated loci. These findings highlight the predominant role played by common variants of modest effect and the diversity of biological mechanisms influencing T2D pathophysiology.

A rare-variant test for high-dimensional data.

Genome-wide association studies have facilitated the discovery of thousands of loci for hundreds of phenotypes. However, the issue of missing heritability remains unsolved for most complex traits. Locus discovery could be enhanced with both improved power through multi-phenotype analysis (MPA) and use of a wider allele frequency range, including rare variants (RVs). MPA methods for single-variant association have been proposed, but given their low power for RVs, more efficient approaches are required. We propose multi-phenotype analysis of rare variants (MARV), a burden test-based method for RVs extended to the joint analysis of multiple phenotypes through a powerful reverse regression technique. Specifically, MARV models the proportion of RVs at which minor alleles are carried by individuals within a genomic region as a linear combination of multiple phenotypes, which can be both binary and continuous, and the method accommodates directly the genotyped and imputed data. The full model, including all phenotypes, is tested for association for discovery, and a more thorough dissection of the phenotype combinations for any set of RVs is also enabled. We show, via simulations, that the type I error rate is well controlled under various correlations between two continuous phenotypes, and that the method outperforms a univariate burden test in all considered scenarios. Application of MARV to 4876 individuals from the Northern Finland Birth Cohort 1966 for triglycerides, high- and low-density lipoprotein cholesterols highlights known loci with stronger signals of association than those observed in univariate RV analyses and suggests novel RV effects for these lipid traits.