Improved imputation of low-frequency and rare variants using the UK10K haplotype reference panel
Huang J., Howie B., McCarthy S., Memari Y., Walter K., Min JL., Danecek P., Malerba G., Trabetti E., Zheng H-F., Al Turki S., Amuzu A., Anderson CA., Anney R., Antony D., Artigas MS., Ayub M., Bala S., Barrett JC., Barroso I., Beales P., Benn M., Bentham J., Bhattacharya S., Birney E., Blackwood D., Bobrow M., Bochukova E., Bolton PF., Bounds R., Boustred C., Breen G., Calissano M., Carss K., Pablo Casas J., Chambers JC., Charlton R., Chatterjee K., Chen L., Ciampi A., Cirak S., Clapham P., Clement G., Coates G., Cocca M., Collier DA., Cosgrove C., Cox T., Craddock N., Crooks L., Curran S., Curtis D., Daly A., Day INM., Day-Williams A., Dedoussis G., Down T., Du Y., van Duijn CM., Dunham I., Edkins S., Ekong R., Ellis P., Evans DM., Farooqi IS., Fitzpatrick DR., Flicek P., Floyd J., Foley AR., Franklin CS., Futema M., Gallagher L., Gasparini P., Gaunt TR., Geihs M., Geschwind D., Greenwood C., Griffin H., Grozeva D., Guo X., Guo X., Gurling H., Hart D., Hendricks AE., Holmans P., Huang L., Hubbard T., Humphries SE., Hurles ME., Hysi P., Iotchkova V., Isaacs A., Jackson DK., Jamshidi Y., Johnson J., Joyce C., Karczewski KJ., Kaye J., Keane T., Kemp JP., Kennedy K., Kent A., Keogh J., Khawaja F., Kleber ME., van Kogelenberg M., Kolb-Kokocinski A., Kooner JS., Lachance G., Langenberg C., Langford C., Lawson D., Lee I., van Leeuwen EM., Lek M., Li R., Li Y., Liang J., Lin H., Liu R., Lönnqvist J., Lopes LR., Lopes M., Luan J., MacArthur DG., Mangino M., Marenne G., März W., Maslen J., Matchan A., Mathieson I., McGuffin P., McIntosh AM., McKechanie AG., McQuillin A., Metrustry S., Migone N., Mitchison HM., Moayyeri A., Morris J., Morris R., Muddyman D., Muntoni F., Nordestgaard BG., Northstone K., O'Donovan MC., O'Rahilly S., Onoufriadis A., Oualkacha K., Owen MJ., Palotie A., Panoutsopoulou K., Parker V., Parr JR., Paternoster L., Paunio T., Payne F., Payne SJ., Perry JRB., Pietilainen O., Plagnol V., Pollitt RC., Povey S., Quail MA., Quaye L., Raymond L., Rehnström K., Ridout CK., Ring S., Ritchie GRS., Roberts N., Robinson RL., Savage DB., Scambler P., Schiffels S., Schmidts M., Schoenmakers N., Scott RH., Scott RA., Semple RK., Serra E., Sharp SI., Shaw A., Shihab HA., Shin S-Y., Skuse D., Small KS., Smee C., Smith GD., Southam L., Spasic-Boskovic O., Spector TD., St Clair D., St Pourcain B., Stalker J., Stevens E., Sun J., Surdulescu G., Suvisaari J., Syrris P., Tachmazidou I., Taylor R., Tian J., Tobin MD., Toniolo D., Traglia M., Tybjaerg-Hansen A., Valdes AM., Vandersteen AM., Varbo A., Vijayarangakannan P., Visscher PM., Wain LV., Walters JTR., Wang G., Wang J., Wang Y., Ward K., Wheeler E., Whincup P., Whyte T., Williams HJ., Williamson KA., Wilson C., Wilson SG., Wong K., Xu C., Yang J., Zaza G., Zeggini E., Zhang F., Zhang P., Zhang W., Gambaro G., Richards JB., Durbin R., Timpson NJ., Marchini J., Soranzo N.
AbstractImputing genotypes from reference panels created by whole-genome sequencing (WGS) provides a cost-effective strategy for augmenting the single-nucleotide polymorphism (SNP) content of genome-wide arrays. The UK10K Cohorts project has generated a data set of 3,781 whole genomes sequenced at low depth (average 7x), aiming to exhaustively characterize genetic variation down to 0.1% minor allele frequency in the British population. Here we demonstrate the value of this resource for improving imputation accuracy at rare and low-frequency variants in both a UK and an Italian population. We show that large increases in imputation accuracy can be achieved by re-phasing WGS reference panels after initial genotype calling. We also present a method for combining WGS panels to improve variant coverage and downstream imputation accuracy, which we illustrate by integrating 7,562 WGS haplotypes from the UK10K project with 2,184 haplotypes from the 1000 Genomes Project. Finally, we introduce a novel approximation that maintains speed without sacrificing imputation accuracy for rare variants.