Masahiro Kanai

I am an Instructor at Dr. Ramnik Xavier Lab at Massachusetts General Hospital & Broad Institute of MIT and Harvard. I completed my Ph.D. in Biomedical Informatics at Harvard University, co-advised by Drs. Mark Daly and Hilary Finucane at the Analytic and Translational Genetics Unit. I received fellowships from the Nakajima Foundation and the Masason Foundation.

As a statistical geneticist, my research focuses on understanding the genetic architecture of complex diseases through cross-population fine-mapping and single-cell molecular QTL mapping. In my current research, I leverage single-nucleus multiome profiling of ~10M PBMCs in FinnGen to comprehensively map caQTLs, eQTLs, and enhancer-gene links, connecting variants to function through regulatory cascades. In my thesis project, I created a comprehensive atlas of putative causal variants from multiple large-scale biobanks to gain insights from complex trait fine-mapping across diverse populations. Prior to joining ATGU, I completed my B.S. degree in Japan, where I worked closely with Dr. Yukinori Okada to study the genetics of complex traits in the Japanese population using the BioBank Japan data.

My curriculum vitae can be found here.

News

Sep 30, 2026	The FinnGen immune multiome atlas (Kanai, M. et al.) is now published in Nature! See also the Broad Institute news.
Sep 18, 2025	The Pan-UKBB flagship manuscript (Karczewski, KJ., Gupta, R., *Kanai, M., et al.) is now published in Nature Genetics!
Sep 6, 2023	Kanai, M., The COVID-19 Host Genetics Initiative. (2023) is now published in Nature!
Nov 5, 2022	Kanai, M. et al. (2022) is now published in Cell Genomics!
Apr 7, 2022	Weissbrod, O., Kanai, M., *Shi, H. et al. (2022) is now published in Nature Genetics!

Selected Publications and Preprints [full list]

Link to Google Scholar | ORCID | PubMed

* denotes equal contribution

Population-scale immune multiome atlas reveals regulatory disease mechanisms

Kanai, M., Delorey, T. M., Honkanen, J., ..., Rodosthenous, R. S., Juvila, J., Murphy, S., Teixeira-Soldano, I., Hwang, H. S., Karjalainen, J., Halonen, J., Panagiotaropoulou, G., Zhang, Y., McCabe, C., Chen, E., Nanki, K., Yoshida, T., Liu, K., Glean, M., Mehrotra, N., Finan, E. P., Chafamo, D., Zhu, Y., Arvas, M., Ruotsalainen, S., Zheng, Z., Tomofuji, Y., Tokuyasu, D., Namba, S., Sonehara, K., BioBank Japan Project., FinnGen., Okada, Y., Reeve, M. P., Kurki, M., Porter, C. B. M., Ashenberg, O., Zhou, W., Pitkänen, K., Partanen, J., Palotie, A., Graham, D. B., Daly, M. J., and Xavier, R. J. [Show fewer authors]

Nature (2026)

Abstract medRxiv Code Data Press

Most disease-associated genetic variants lie in non-coding regions1,2, yet mechanistic insights are limited by the lack of an empirical framework for characterizing the molecular consequences of regulatory variation. Single-cell molecular quantitative trait locus (QTL) mapping3,4 connects variants to gene regulation but lacks the power and simultaneous measurements to trace mechanisms from chromatin to expression5. Here we show that population-scale simultaneous profiling of chromatin accessibility and gene expression across immune cells reveals regulatory architectures connecting variants to disease. From paired single-nucleus assay for transposase-accessible chromatin-sequencing (snATAC–seq) and single-nucleus RNA-sequencing (snRNA-seq) analysis of 10 million peripheral blood mononuclear cells in 1,108 Finnish individuals6, we identify 51,083 cis-expression QTLs for 20,829 genes, 338,100 cis-chromatin accessibility QTLs for 210,584 peaks, 119,094 putative causal variants and 593,765 peak–gene links. Variants completing chromatin-to-expression cascades show twice the disease colocalization of chromatin-only effects, with massively parallel reporter assays7 validating 10,428 fine-mapped molecular QTLs. At evolutionarily constrained genes, we identify multilayered regulatory buffering, in which chromatin accessibility changes occur with normal effect sizes, but transmission to expression is attenuated through weaker, more numerous enhancer–gene links. This reconciles why disease variants preferentially target constrained genes despite apparent expression QTL depletion8–11. Analysis using a massively parallel reporter assay7 confirms that this buffering acts downstream of the regulatory element, with constraint operating at the chromatin-to-expression interface rather than on intrinsic cis-regulatory activity. Our atlas provides testable hypotheses for over half of immune disease associations, illustrated by cascades at autoimmune loci (TICAM1 and RHOH) and Finnish-enriched variants (TNRC18 and IL21R). Population-scale simultaneous profiling of chromatin accessibility and gene expression across immune cells reveals regulatory architectures that connect genetic variants to disease.
sc-pcQTL: hurdle-based co-expression modeling for multi-gene QTL mapping in single-cell RNA-seq data

Zhang, J., Huang, Y., Claussnitzer, M., *Kanai, M., and *Zhou, W.

bioRxiv (2026)

Abstract PDF Code

Motivation: Single-cell expression quantitative trait locus (eQTL) studies can resolve cell-type-specific genetic effects, but conventional gene-by-gene analyses do not directly capture coordinated genetic regulation of neighboring genes. Principal-component QTL (pcQTL) mapping can summarize such multi-gene effects, but existing approaches were developed for bulk expression and are not designed for sparse single-cell counts. Results: We developed sc-pcQTL, a framework that applies two-component hurdle modeling and sliding-window clustering to identify local co-expression clusters, summarizes each cluster using principal components, and maps cis -pcQTLs. In simulations, the individual hurdle components controlled type I error, while the component-union screening rule was substantially more powerful than donor-level pseudobulk correlation tests. Applied to 1.24 million peripheral blood mononuclear cells from 982 OneK1K donors across 10 cell types, sc-pcQTL identified 2,485 local co-expression clusters and conducted QTL mapping for 4,353 cluster-PC phenotypes at single-cell resolution, of which 2,040 had at least one significant cis -pcQTL association. Fine-mapping and colocalization with genome-wide association study loci across 1,163 phenotypes in the FinnGen study identified 394 colocalized QTL-GWAS signal groups. Each group comprised fine-mapped QTL and GWAS signals connected through one or more colocalization links within the same cell type and local gene cluster. Of these groups, 46 were pcQTL-specific and contained no colocalized single-gene eQTL from a constituent gene. Locus-level analyses further revealed cell-type-specific multi-gene regulatory effects. Thus, sc-pcQTL complements conventional single-gene eQTL analysis by identifying trait-relevant regulatory signals shared across neighboring genes.
Pan-UK Biobank genome-wide association analyses enhance discovery and resolution of ancestry-enriched effects

*Karczewski, K. J., *Gupta, R., *Kanai, M., ..., Lu, W., Tsuo, K., Wang, Y., Walters, R. K., Turley, P., Callier, S., Shah, N. N., Baya, N., Palmer, D. S., Goldstein, J. I., Sarma, G., Solomonson, M., Cheng, N., Bryant, S., Churchhouse, C., Cusick, C. M., Poterba, T., Compitello, J., King, D., Zhou, W., Seed, C., Finucane, H. K., Daly, M. J., Neale, B. M., Atkinson, E. G., and Martin, A. R. [Show fewer authors]

Nature Genetics 57, 2408–2417 (2025)

Abstract PDF PubMed medRxiv Code Website

Large biobanks, such as the UK Biobank (UKB), enable massive phenome by genome-wide association studies that elucidate genetic etiology of complex traits. However, people from diverse genetic ancestry groups are often excluded from association analyses due to concerns about population structure introducing false positive associations. Here we generate mixed model associations and meta-analyses across genetic ancestry groups, inclusive of a larger fraction of the UK Biobank than previous efforts, to produce freely available summary statistics for 7,266 traits. We build a quality control and analysis framework informed by genetic architecture. Overall, we identify 14,676 significant loci (P < 5 × 10−8) in the meta-analysis that were not found in the EUR genetic ancestry group alone, including new associations, for example between CAMK2D and triglycerides. We also highlight associations from ancestry-enriched variation, including a known pleiotropic missense variant in G6PD associated with several biomarker traits. We release these results publicly alongside frequently asked questions that describe caveats for interpretation of results, enhancing available resources for interpretation of risk variants across diverse populations. Genome-wide analyses for 7,266 traits leveraging data from several genetic ancestry groups in UK Biobank identify new associations and enhance resources for interpreting risk variants across diverse populations.
A second update on mapping the human genetic architecture of COVID-19

Kanai, M., and The COVID-19 Host Genetics Initiative.

Nature 621, E7–E26 (2023)

Lead author of this report; one of the writing group leads for all three of the Initiative's Nature reports (2021, 2022, 2023)

Abstract PDF PubMed medRxiv Website

Investigating the role of host genetic factors in COVID-19 severity and susceptibility can inform our understanding of the underlying biological mechanisms that influence adverse outcomes and drug development. Here we present a second updated genome-wide association study (GWAS) on COVID-19 severity and infection susceptibility to SARS-CoV-2 from the COVID-19 Host Genetic Initiative (data release 7). We performed a meta-analysis of up to 219,692 cases and over 3 million controls, identifying 51 distinct genome-wide significant loci—adding 28 loci from the previous data release. The increased number of candidate genes at the identified loci helped to map three major biological pathways that are involved in susceptibility and severity: viral entry, airway defence in mucus and type I interferon.
Meta-analysis fine-mapping is often miscalibrated at single-variant resolution

Kanai, M., Elzur, R., Zhou, W., ..., Global Biobank Meta-analysis Initiative., Daly, M. J., and Finucane, H. K. [Show fewer authors]

Cell Genomics 2, 100210 (2022)

Abstract PDF PubMed medRxiv Code Website

Meta-analysis is pervasively used to combine multiple genome-wide association studies (GWASs). Fine-mapping of meta-analysis studies is typically performed as in a single-cohort study. Here, we first demonstrate that heterogeneity (e.g., of sample size, phenotyping, imputation) hurts calibration of meta-analysis fine-mapping. We propose a summary statistics-based quality-control (QC) method, suspicious loci analysis of meta-analysis summary statistics (SLALOM), that identifies suspicious loci for meta-analysis fine-mapping by detecting outliers in association statistics. We validate SLALOM in simulations and the GWAS Catalog. Applying SLALOM to 14 meta-analyses from the Global Biobank Meta-analysis Initiative (GBMI), we find that 67% of loci show suspicious patterns that call into question fine-mapping accuracy. These predicted suspicious loci are significantly depleted for having nonsynonymous variants as lead variant (2.7\times; Fisher’s exact p = 7.3 \times 10−4). We find limited evidence of fine-mapping improvement in the GBMI meta-analyses compared with individual biobanks. We urge extreme caution when interpreting fine-mapping results from meta-analysis of heterogeneous cohorts.
Leveraging fine-mapping and multipopulation training data to improve cross-population polygenic risk scores

*Weissbrod, O., *Kanai, M., *Shi, H., ..., Gazal, S., Peyrot, W. J., Khera, A. V., Okada, Y., The Biobank Japan Project., Martin, A. R., Finucane, H. K., and Price, A. L. [Show fewer authors]

Nature Genetics 54, 450–458 (2022)

Abstract PDF PubMed medRxiv Code Data

Polygenic risk scores suffer reduced accuracy in non-European populations, exacerbating health disparities. We propose PolyPred, a method that improves cross-population polygenic risk scores by combining two predictors: a new predictor that leverages functionally informed fine-mapping to estimate causal effects (instead of tagging effects), addressing linkage disequilibrium differences, and BOLT-LMM, a published predictor. When a large training sample is available in the non-European target population, we propose PolyPred+, which further incorporates the non-European training data. We applied PolyPred to 49 diseases/traits in four UK Biobank populations using UK Biobank British training data, and observed relative improvements versus BOLT-LMM ranging from +7% in south Asians to +32% in Africans, consistent with simulations. We applied PolyPred+ to 23 diseases/traits in UK Biobank east Asians using both UK Biobank British and Biobank Japan training data, and observed improvements of +24% versus BOLT-LMM and +12% versus PolyPred. Summary statistics-based analogs of PolyPred and PolyPred+ attained similar improvements. PolyPred and PolyPred+ methods that leverage fine-mapping and non-European training data significantly improve cross-population polygenic prediction accuracy when applied to diseases and complex traits in UK Biobank populations.
Insights from complex trait fine-mapping across diverse populations

Kanai, M., Ulirsch, J. C., Karjalainen, J., ..., Kurki, M., Karczewski, K. J., Fauman, E. B., Wang, Q. S., Jacobs, H., Aguet, F., Ardlie, K. G., Kerimov, N., Alasoo, K., Benner, C., Ishigaki, K., Sakaue, S., Reilly, S., The BioBank Japan Project., FinnGen., Kamatani, Y., Matsuda, K., Palotie, A., Neale, B. M., Tewhey, R., Sabeti, P. C., Okada, Y., Daly, M. J., and Finucane, H. K. [Show fewer authors]

medRxiv (2021)

Abstract PDF Code Data Website

Despite the great success of genome-wide association studies (GWAS) in identifying genetic loci significantly associated with diseases, the vast majority of causal variants underlying disease-associated loci have not been identified. To create an atlas of causal variants, we performed and integrated fine-mapping across 148 complex traits in three large-scale biobanks (BioBank Japan, FinnGen, and UK Biobank; total n = 811,261), resulting in 4,518 variant-trait pairs with high posterior probability (> 0.9) of causality. Of these, we found 285 high-confidence variant-trait pairs replicated across multiple populations, and we characterized multiple contributors to the surprising lack of overlap among fine-mapping results from different biobanks. By studying the bottlenecked Finnish and Japanese populations, we identified 21 and 26 putative causal coding variants with extreme allele frequency enrichment (> 10-fold) in these two populations, respectively. Aggregating data across populations enabled identification of 1,492 unique fine-mapped coding variants and 176 genes in which multiple independent coding variants influence the same trait (i.e., with an allelic series of coding variants). Our results demonstrate that fine-mapping in diverse populations enables novel insights into the biology of complex traits by pinpointing high-confidence causal variants for further characterization.
A cross-population atlas of genetic associations for 220 human phenotypes

*Sakaue, S., *Kanai, M., Tanigawa, Y., ..., Karjalainen, J., Kurki, M., Koshiba, S., Narita, A., Konuma, T., Yamamoto, K., Akiyama, M., Ishigaki, K., Suzuki, A., Suzuki, K., Obara, W., Yamaji, K., Takahashi, K., Asai, S., Takahashi, Y., Suzuki, T., Shinozaki, N., Yamaguchi, H., Minami, S., Murayama, S., Yoshimori, K., Nagayama, S., Obata, D., Higashiyama, M., Masumoto, A., Koretsune, Y., Ito, K., Terao, C., Yamauchi, T., Komuro, I., Kadowaki, T., Tamiya, G., Yamamoto, M., Nakamura, Y., Kubo, M., Murakami, Y., Yamamoto, K., Kamatani, Y., Palotie, A., Rivas, M. A., Daly, M. J., Matsuda, K., and Okada, Y. [Show fewer authors]

Nature Genetics 53, 1415–1424 (2021)

Featured on the cover of Nature Genetics

Abstract PDF PubMed medRxiv Website Press

Current genome-wide association studies do not yet capture sufficient diversity in populations and scope of phenotypes. To expand an atlas of genetic associations in non-European populations, we conducted 220 deep-phenotype genome-wide association studies (diseases, biomarkers and medication usage) in BioBank Japan (n = 179,000), by incorporating past medical history and text-mining of electronic medical records. Meta-analyses with the UK Biobank and FinnGen (ntotal = 628,000) identified 5,000 new loci, which improved the resolution of the genomic map of human traits. This atlas elucidated the landscape of pleiotropy as represented by the major histocompatibility complex locus, where we conducted HLA fine-mapping. Finally, we performed statistical decomposition of matrices of phenome-wide summary statistics, and identified latent genetic components, which pinpointed responsible variants and biological mechanisms underlying current disease classifications across populations. The decomposed components enabled genetically informed subtyping of similar diseases (for example, allergic diseases). Our study suggests a potential avenue for hypothesis-free re-investigation of human diseases through genetics. Genome-wide analyses in BioBank Japan, UK Biobank and FinnGen identify 5,000 new loci associated with 220 human traits. Statistical decomposition of matrices of phenome-wide summary statistics further highlights variants underpinning diseases across populations.
Trans-biobank analysis with 676,000 individuals elucidates the association of polygenic risk scores of complex traits with human lifespan

*Sakaue, S., *Kanai, M., Karjalainen, J., ..., Akiyama, M., Kurki, M., Matoba, N., Takahashi, A., Hirata, M., Kubo, M., Matsuda, K., Murakami, Y., Daly, M. J., Kamatani, Y., and Okada, Y. [Show fewer authors]

Nature Medicine 26, 542–548 (2020)

Featured in News & Views of Nature Medicine

Abstract PDF PubMed bioRxiv Press

While polygenic risk scores (PRSs) are poised to be translated into clinical practice through prediction of inborn health risks1, a strategy to utilize genetics to prioritize modifiable risk factors driving heath outcome is warranted2. To this end, we investigated the association of the genetic susceptibility to complex traits with human lifespan in collaboration with three worldwide biobanks (ntotal = 675,898; BioBank Japan (n = 179,066), UK Biobank (n = 361,194) and FinnGen (n = 135,638)). In contrast to observational studies, in which discerning the cause-and-effect can be difficult, PRSs could help to identify the driver biomarkers affecting human lifespan. A high systolic blood pressure PRS was trans-ethnically associated with a shorter lifespan (hazard ratio = 1.03[1.02-1.04], Pmeta = 3.9 \times 10-13) and parental lifespan (hazard ratio = 1.06[1.06-1.07], P = 2.0 \times 10-86). The obesity PRS showed distinct effects on lifespan in Japanese and European individuals (Pheterogeneity = 9.5 \times 10-8 for BMI). The causal effect of blood pressure and obesity on lifespan was further supported by Mendelian randomization studies. Beyond genotype-phenotype associations, our trans-biobank study offers a new value of PRSs in prioritization of risk factors that could be potential targets of medical treatment to improve population health.
Genetic analysis of quantitative traits in the Japanese population links cell types to complex human diseases

Kanai, M., Akiyama, M., Takahashi, A., ..., Matoba, N., Momozawa, Y., Ikeda, M., Iwata, N., Ikegawa, S., Hirata, M., Matsuda, K., Kubo, M., Okada, Y., and Kamatani, Y. [Show fewer authors]

Nature Genetics 50, 390–400 (2018)

Abstract PDF PubMed Press

Clinical measurements can be viewed as useful intermediate phenotypes to promote understanding of complex human diseases. To acquire comprehensive insights into the underlying genetics, here we conducted a genome-wide association study (GWAS) of 58 quantitative traits in 162,255 Japanese individuals. Overall, we identified 1,407 trait-associated loci (P < 5.0 \times 10-8), 679 of which were novel. By incorporating 32 additional GWAS results for complex diseases and traits in Japanese individuals, we further highlighted pleiotropy, genetic correlations, and cell-type specificity across quantitative traits and diseases, which substantially expands the current understanding of the associated genetics and biology. This study identified both shared polygenic effects and cell-type specificity, represented by the genetic links among clinical measurements, complex diseases, and relevant cell types. Our findings demonstrate that even without prior biological knowledge of cross-phenotype relationships, genetics corresponding to clinical measurements successfully recapture those measurements’ relevance to diseases, and thus can contribute to the elucidation of unknown etiology and pathogenesis.