BOLT-LMM v2.0 User Manual
Contents
1. bolt which we have tested on several Linux systems We strongly recommend using this static executable because it is well optimized and no further installation is required If you wish to compile your own version of the BOLT LMM software from the source code in the src subdirectory provided in a separate download licensed under GNU GPLv3 you will need to ensure that library dependencies are fulfilled and will need to make appropriate modifica tions to the Makefile e Library dependencies BLAS LAPACK numerical libraries The speed of the BOLT LMM software depends critically on the efficiency of the BLAS LAPACK implementation it is linked against We recommend the Intel Math Kernel Library MKL if available otherwise ATLAS may be a good alternative Boost C libraries BOLT LMM links against the Boost program_options and iostreams libraries which need to be installed after downloading and unzipping Boost NLopt numerical optimization library 12 e Makefile Paths to libraries need to be modified appropriately Note that the released ver sion of the Makefile does not set the flag DUSE_MKL_MALLOC This flag turns on the Intel MKL s fast memory manager replacing calls to_mm_malloc with mk1_malloc which may improve memory performance but we have observed crashes on some systems when using mk 1_malloc For reference the provided bolt executable was created on the Harvard Medical School Orches tra re2. PLINK to LD prune to 500K SNPs via indep pairwise 50 5 r2thresh for an appropriate r2thresh 4 Run BOLT LMM using the final hard called SNPs as the bfile or bed bim fam argument specifying the imputed SNPs as additional association test SNPs using one of the formats below Imputed SNPs in dosage format This input format consists of one or more dosageFile parameters specifying files that contain real valued genotype expectations at imputed SNPs Each line of a dosageFile should be formatted as follows rsID chr pos allelel allele0 dosag E allelel x N Missing i e uncalled dosages can be specified with 9 You will also need to provide one additional dosageFidlidFile specifying the PLINK FIDs and IIDs of samples that the dosages correspond to See the example subdirectory for an example Imputed SNPs in IMPUTE2 format You may also specify imputed SNPs as output by the IMPUTE2 software 15 The IMPUTE2 genotype file format is as follows snpID rsID pos allelel allele0 p 11 p 10 p 00 x N BOLT LMM ignores the snpID field Here instead of dosages each genotype entry contains in dividual probabilities of the individual being homozygous for allele1 heterozygous and homozy gous for alleleO The three probabilities need not sum to 1 allowing for genotype uncertainty if the sum of the probabilities is less than the impute2CallThresh parameter BOLT LMM treats the genoty
3. be used must be specified using either a covarCol for categorical covariates or a qCovarCol for quantitative covariates option Categorical covariate values are allowed to be any text strings not containing whitespace each unique text string in a column corresponds to a category Quantitative covariate values must be numeric with the exception of NA In either case values of 9 and NA are interpreted as missing data If groups of covariates of the same type are numbered sequentially they may be specified using array shorthand e g qCovarCol PC 1 10 for columns PC1 PC2 PC10 5 4 Missing data treatment Individuals with missing phenotypes are ignored By default individuals with any missing co variates are also ignored this approach is commonly used and referred to as complete case analysis As an alternative we have also implemented the missing indicator method via the covarUseMissingIndic option which adds indicator variables demarcating missing sta tus as additional covariates Missing genotypes are replaced with per SNP averages 5 5 Genotype QC BOLI LMM and BOLT REML automatically filter SNPs and individuals with missing rates ex ceeding thresholds of 0 1 These thresholds may be modified using maxMissingPerSnp and maxMissingPerlIndiv Note that filtering is not performed based on minor allele frequency or deviation from Hardy Weinberg equilibrium Allele frequency and missin
4. faster than existing methods both when using the Bayesian mix ture model and when specialized to standard mixed model association BOLT LMM is described in ref 1 Loh P R Tucker G Bulik Sullivan BK Vilhj lmsson BJ Finucane HK Salem RM Chasman DI Ridker PM Neale BM Berger B Patterson N and Price AL Efficient Bayesian mixed model analysis increases association power in large cohorts Nature Genetics 2015 1 2 BOLT REML variance components analysis The BOLT REML algorithm estimates heritability explained by genotyped SNPs and genetic cor relations among multiple traits measured on the same set of individuals Like the GCTA soft ware 10 BOLT REML applies variance components analysis to perform these tasks supporting both multi component modeling to partition SNP heritability and multi trait modeling to estimate correlations BOLI REML applies a Monte Carlo algorithm that is much faster than standard methods for variance components analysis e g GCTA at large sample sizes BOLI REML is described in ref 11 Loh P R Bhatia G Gusev A Finucane HK Bulik Sullivan BK Pollack SJ PGC SCZ Working Group de Candia TR Lee SH Wray NR Kendler KS O Donovan MC Neale BM Patterson N and Price AL Contrasting regional architectures of schizophrenia and other complex diseases using fast variance components analysis bioRxiv 2015 2 Installation We provide a standalone i e statically linked 64 bit Linux executable
5. 7 1 Multiple variance components 4 a 11 hed Multiple traits ae oi e ais de e ee a e rr ee Se a 11 7 3 Initial variance parameter guesses 2 2 ear ea eRe cres ee ee 11 7 4 Trading a little accuracy for speed aoe an esca a 11 8 Output 12 8 1 BOLT LMM association test statistics o 0 00000 eee eee 12 8 2 BOLTI REML output and logging 12 9 Changelog 13 10 Website and contact info 13 11 License 13 1 Overview The BOLT LMM software package currently consists of two main algorithms the BOLT LMM algorithm for mixed model association testing and the BOLT REML algorithm for variance com ponents analysis i e partitioning of SNP heritability and estimation of genetic correlations 1 1 BOLT LMM mixed model association testing The BOLT LMM algorithm computes statistics for testing association between phenotype and genotypes using a linear mixed model LMM 1 By default BOLT LMM assumes a Bayesian mixture of normals prior for the random effect attributed to SNPs other than the one being tested This model generalizes the standard infinitesimal mixed model used by existing mixed model as sociation methods e g EMMAX 2 FaST LMM 3 6 GEMMA 7 GRAMMAR Gamma 8 GCTA LOCO 9 providing an opportunity for increased power to detect associations while con trolling false positives Additionally BOLT LMM applies algorithmic advances to compute mixed model association statistics much
6. BOLT LMM v2 0 User Manual Po Ru Loh March 13 2015 Contents 1 Overview 1 1 BOLT LMM mixed model association testing 2 1 2 BOLT REML variance components analysis o e 2 Installation 2 1 Running BOLT LMM and BOLEREML o o 2 2 A A RR Ded OD ae eG a o rt a a a a a aa a a a ae 3 Computing requirements 3 1 Operating system a e a IAEA 32a MEM A A AA ee A Deo RUINA ME a A a o E e e A A A a areas Bg A Ree BK 4 Input output file naming conventions 4 1 Automatic gzip de Jcompression o a 4 2 Arrays of input files and covariates a 5 Input Deli En Pess A A NA aga en a i en 5 1 1 Reference genetic maps La a a ol a Bo eas es S 5 1 2 Imputed SNP dosages be a a e Ee eee E e 5 2 PNETIOLY DOS 2 cai dc a ee we eee Dua EON Al IS rele Mo She Hed ee She E a oe Se te Soe te Sm te Soe ee Se te 5 4 Missing data treatment ra ii BIA Gk A he uk aa eee ac 2 Y SASOROLY pE QG yi ge i i i cic Bee Erk ha ap ck et Ses ew BS ed IIA Boome 2 5 6 User specified filters 21 5 2 2 bo bb ee ae RP Ee ee ee ee Ee 6 Association analysis BOLT LMM 9 6 1 Mixed model association tests 2 2 0 0 002 eee eee ee 9 6 2 BOLT LMM mixed model association options 20 4 10 6 2 1 Restricting SNPs used in the mixed model 10 6 3 Standard linear regression 5 4 A gs 10 7 Variance components analysis BOLT REML 10
7. ar regression chi square statistics and p values in additional output columns CHISQ_LINREG and P_LINREG Note that unlike mixed model association linear regression is susceptible to population stratification so you may wish to include principal components computed using other software e g PLINK2 or FastPCA 16 in EIGENSOFT v6 0 as covariates when performing linear regression 7 Variance components analysis BOLT REML Using the rem1 option invokes the BOLT REML algorithm for estimating heritability parame ters and genetic correlations 10 7 1 Multiple variance components To assign SNPs to different variance components specify a modelSnps file in which each whitespace delimited line contains a SNP ID typically an rs number followed by the name of the variance component to which it belongs 7 2 Multiple traits To perform multi trait variance components analysis specify multiple phenoCol parameter value flags corresponding to different columns in the same phenoFile BOLT REML cur rently only supports multi trait analysis of traits phenotyped on a single set of individuals so any individuals with at least one missing phenotype will be ignored For D traits BOLT REML esti mates D heritability parameters per variance component and D D 1 2 correlations per variance component including the residual variance component 7 3 Initial variance parameter guesses To specify a set of variance paramet
8. ers at which to start REML iteration which may save time compared to the default procedure used by BOLT REML if you have good initial guesses use remlGuessStr string with the following format For each variance component start ing with the residual term which is automatically named env noise specify the name of the variance component followed by the initial guess For instance a model with two non residual variance components named vel and vc2 in the mode1Snps file could have variance param eter guesses specified by remlGuessStr env noise 0 5 vel 0 2 vc2 0 3 Note that the sum of the estimates must equal 1 BOLT REML will automatically normalize the phenotype accordingly For multi trait analysis of D traits the remlGuessStr needs to specify both guesses of D variance proportions and D D 1 2 pairwise correlations per variance component Viewing these values as entries of an upper triangular matrix with variance proportions on the diagonal and correlations above the diagonal you should specify these D D 1 2 values after each variance component name by reading them off left to right top to bottom 7 4 Trading a little accuracy for speed BOLI REML uses a Monte Carlo algorithm to increase REML optimization speed 11 By de fault BOLT REML performs an initial optimization using 15 Monte Carlo trials and then refines parameter estimates using 100 Monte Carlo trials If computational cost is a concern or to perf
9. gness of each SNP are included in the BOLT LMM association test output however and we recommend checking these values and Hardy Weinberg p values which are easily computed using PLINK2 hardy when following up on significant associations 5 6 User specified filters Individuals to remove from the analysis may be specified in one or more remove files listing FID and IIDs one individual per line Similarly SNPs to exclude from the analysis may be specified in one ore more exclude files listing SNP IDs typically rs numbers 6 Association analysis BOLT LMM 6 1 Mixed model association tests BOLT LMM computes two association statistics X BoLT LMM and X BOLT LMMinf described in detail in our manuscript 1 e BOLT LMM Association test on residuals from Bayesian modeling using a mixture of normals prior on SNP effect sizes This approach can fit non infinitesimal traits with loci having moderate to large effects allowing increased association power e BOLT LMM inf Standard infinitesimal mixed model association This statistic ap proximates the standard approach used by existing software 6 2 BOLT LMM mixed model association options The BOLI LMM software offers the following options for mixed model analysis e 1mm Performs default BOLI LMM analysis which consists of 1a estimating heritabil ity parameters 1b computing the BOLT LMM inf statistic 2a estimating Gaussian m
10. ix ture parameters and 2b computing the BOLT LMM statistic only if an increase in power is expected If BOLT LMM determines based on cross validation that the non infinitesimal model is likely to yield no increase in power the BOLT LMM Bayesian mixed model statistic is not computed e lmmInfOnly Computes only infinitesimal mixed model association statistics 1 e steps la and 1b e lmmForceNonInf Computes both the BOLIT LMM inf and BOLT LMM statistics re gardless of whether or not an increase in power is expected from the latter 6 2 1 Restricting SNPs used in the mixed model If millions of SNPs are available from imputation we suggest including lt 1 million SNPs at a time in the mixed model using the mode1Snps option when performing association analy sis Using an LD pruned set of lt 1 million SNPs should achieve near optimal power and correction for confounding while reducing computational cost and improving convergence Note that even when a file of modelSnps is specified all SNPs in the genotype data are still tested for as sociation only the random effects in the mixed model are restricted to the modelSnps Also note that BOLI LMM automatically performs leave one chromosome out LOCO analysis leav ing out SNPs from the chromosome containing the SNP being tested in order to avoid proximal contamination 4 9 6 3 Standard linear regression Setting the verboseStats flag will output standard line
11. orm exploratory analyses you can skip the refinement step using the remlNoRefine flag This option typically gives 2 3x speedup at the cost of 1 03x higher standard errors 11 8 Output 8 1 BOLT LMM association test statistics BOLI LMM association statistics are output in a tab delimited statsFile file with the fol lowing fields one line per SNP e SNP rs number or ID string e CHR chromosome e BP physical base pair position e GENPOS genetic position either from bim file or interpolated from genetic map e ALLELEL first allele in bim file usually the minor allele used as the effect allele e ALLELEO second allele in bim file used as the reference allele e A1FREO frequency of first allele e F_MISS fraction of individuals with missing genotype at this SNP e BETA effect size from BOLI LMM approximation to infinitesimal mixed model e SE standard error of effect size e P_BOLT_LMM_INF infinitesimal mixed model association test p value P_BOLT_LMM non infinitesimal mixed model association test p value Optional additional output To output chi square statistics for all association tests set the verboseStats flag 8 2 BOLT REML output and logging BOLT REML output i e variance parameter estimates and standard errors is simply printed to the terminal st dout when analysis finishes Both BOLT LMM and BOLT REML write output to stdout and stderr as analysis proceeds we recommend saving this
12. output If you wish to save this output while simultaneously viewing it on the command line you may do so using bolt List of options 2 61 tee cutput Log 12 9 Change log e Version 2 0 Mar 13 2015 Added BOLT REML algorithm for estimating heritability pa rameters Fixed parameter initialization bug that prevented BOLT LMM from running on some systems Implemented various minor improvements to parameter checking e Dec 8 2014 Licensed source code under GPLv3 e Version 1 2 Nov 4 2014 Added support for testing imputed SNPs in 2 dosage format Ricopili plink2 format 2 Fixed bug causing nan heritability estimates e Version 1 1 Oct 17 2014 Added support for testing imputed SNPs with probabilistic dosages e Version 1 0 Aug 8 2014 Initial release 10 Website and contact info Software updates will be posted at the following website http ww hsph harvard edu alkes price software If you have comments or questions about the BOLT LMM software please contact Po Ru Loh loh hsph harvard edu 11 License e The BOLT LMM source code is free software under the GNU General Public License v3 0 GPLv3 e The BOLT LMM executable is freely available but not open source because it was built with the proprietary Intel Math Kernel Library under a non commercial license Specific license terms are as follows Copyright 2014 Harvard University All rights reserved This software is sup plied without an
13. pe as missing To compute association statistics at a list of files containing IMPUTE2 SNPs you may list the files within a impute2FileList file Each line of this file should contain two entries a chromosome number followed by an IMPUTE2 genotype file containing SNPs from that chro mosome You will also need to provide one additional impute2FidlidFile specifying the PLINK FIDs and IDs of samples that the IMPUTE2 genotypes correspond to See the example subdirectory for an example Imputed SNPs in 2 dosage format You may also specify imputed SNPs as output by the Ri copili pipeline and plink2 dosage format 2 This file format consists of file pairs 1 PLINK map files containing information about SNP locations and 2 genotype probability files in the 2 dosage format which consists of a header line SNP Al A2 FID IID x N followed by one line per SNP in the format rsID allelel allele0 p 11 p 10 x N The third genotype probability for each entry is assumed to be p 00 1 p 11 p 10 unlike with the IMPUTE2 format To compute association statistics at SNPs in a list of 2 dosage files you may list the files within a dosage2FileList file Each line of this file should contain two entries a PLINK map file followed by the corresponding genotype file containing probabilities for those SNPs As usual if either file ends with gz it is automatically unzipped otherwise it is assumed to be plain text See the example
14. riates can be specified by the shorthand i j For example data chr 1 22 bim is interpreted as the list of files data chrl bim data chr2 bim data chr22 bim 5 Input 5 1 Genotypes The BOLT LMM software takes genotype input in PLINK 13 binary format bed bim fam For file conversion and data manipulation in general we highly recommend the PLINK2 soft ware 14 which is providing a comprehensive much more efficient update to PLINK If all genotypes are contained in a single bed bim fam file triple with the same file prefix you may simply use the command line option bfile prefix Genotypes may also be split into multiple bed and bim files containing consecutive sets of SNPs e g one bed bim file pair per chromosome either by using multiple bed and bim invocations or by using the file array shorthand described above e g bim data chr 1 22 bim 5 1 1 Reference genetic maps The BOLT LMM package includes reference maps that you can use to interpolate genetic map coordinates from SNP physical base pair positions in the event that your PLINK bin file does not contain genetic coordinates in units of morgans The BOLT LMM association testing algo rithm uses genetic positions to prevent proximal contamination BOLT REML does not use this information To use a reference map use the option geneticMapFile tables genetic_map_hg txt gz selecting the build hg17 hg18 or hg19 corresponding to the ph
15. search computing cluster using Intel Composer XE 12 0 5 MKL 10 3 5 and the Boost 1 42 and NLopt 2 4 2 libraries by invoking make cluster orchestra linking static 2 1 Running BOLT LMM and BOLT REML To run the bolt executable simply invoke bolt on the Linux command line within the BOLT LMM install directory with parameters in the format opt ionName optionValue 2 2 Examples The example subdirectory contains a bash script run_example sh that demonstrates basic use of BOLT LMM on a small example data set Likewise run_example_rem12 sh demon strates BOLT REML e A minimal BOLT LMM invocation looks like bolt bfile geno phenoFile pheno tat phenoCol phenoName lmm LDscoresFile tables LDSCORE 1000G_EUR tab gz statsFile stats tab e A minimal BOLT REML invocation looks like bolt bfile geno phenoFile pheno txt phenoCol phenoName reml modelSnps modelSnps tat To perform multi trait BOLT REML i e estimate genetic correlations provide multiple phenoCol phenoN arguments 2 3 Help To get a list of basic options run bolt h To get a complete list of basic and advanced options run bolt helpFull 3 Computing requirements 3 1 Operating system At the current time we have only compiled and tested BOLT LMM on Linux computing envi ronments however the source code is available if you wish to try compiling BOLELMM for a different operating system 3 2 Memory For typical data
16. sets M N gt 10 000 BOLI LMM and BOLT REML use approximately MN 4 bytes of memory where M is the number of SNPs and N is the number of individuals More precisely e M of SNPs in bin file s that satisfy all of the conditions not listed in any exclude file passed QC filter for missingness listed in modelSnps file s if specified e N of individuals in fam file and not listed in any remove file but pre QC i e N includes individuals filtered due to missing genotypes or covariates 3 3 Running time In practice BOLT LMM and BOLT REML have running times that scale roughly with MN Our largest analyses of real data M 600K SNPs N 60K individuals took 1 day using a single computational core We have also tested BOLT LMM on simulated data sets containing up to N 480K individuals for more details please see the BOLT LMM manuscript 1 3 3 1 Multi threading On multi core machines running time can be reduced by invoking multi threading using the numThreads option 4 Input output file naming conventions 4 1 Automatic gzip de compression The BOLT LMM software assumes that input files ending in gz are gzip compressed and auto matically decompresses them on the fly 1 e without creating a temporary file Similarly BOLT LMM writes gzip compressed output to any output file ending in gz 4 2 Arrays of input files and covariates Arrays of sequentially numbered input files and cova
17. sociation studies Nature Genetics 44 821 824 2012 Svishcheva G R Axenovich T I Belonogova N M van Duijn C M amp Aulchenko Y S Rapid variance components based method for whole genome association analysis Nature Genetics 2012 Yang J Zaitlen N A Goddard M E Visscher P M amp Price A L Advantages and pitfalls in the application of mixed model association methods Nature Genetics 46 100 106 2014 Yang J Lee S H Goddard M E amp Visscher P M GCTA a tool for genome wide complex trait analysis American Journal of Human Genetics 88 76 82 2011 P R L et al Contrasting regional architectures of schizophrenia and other complex diseases using fast variance components analysis bioRxiv 2015 Johnson S G The NL opt nonlinear optimization package URL http ab initio mit edu nlopt Purcell S ef al PLINK a tool set for whole genome association and population based linkage analyses American Journal of Human Genetics 81 559 575 2007 Chang C C et al Second generation PLINK rising to the challenge of larger and richer datasets GigaScience 2015 Howie B N Donnelly P amp Marchini J A flexible and accurate genotype imputation method for the next generation of genome wide association studies PLoS Genetics 5 e1000529 2009 14 16 Galinsky K J et al Fast PCA of very large samples in linear time Abstract presented at the 64th Annual Mee
18. subdirectory for an example 5 2 Phenotypes Phenotypes may be specified in either of two ways e phenoUseFam This option tells BOLELMM and BOLT REML to use the last 6th column of the fam file as the phenotypes This column must be numeric so case control phenotypes should be 0 1 coded and missing values should be indicated with 9 e phenoFile and phenoCol Alternatively phenotypes may be provided in a sep arate whitespace delimited file specified with phenoFile with the first line contain ing column headers and subsequent lines containing records one per individual The first two columns must be FID and IID the PLINK identifiers of an individual Any number of columns may follow the column containing the phenotype to analyze is specified with phenoCol Values of 9 and NA are interpreted as missing data All other values in the column should be numeric The records in lines following the header line need not be in sorted order and need not match the individuals in the genotype data i e fam file BOLT LMM and BOLT REML will analyze only the individuals in the intersection of the genotype and phenotype files and will output a warning if these sets do not match 5 3 Covariates Covariate data may be specified in a file covarF ile with the same format as the alternate phenotype file described above The same file may be used for both phenotypes and covari ates Each covariate to
19. ting of The American Society of Human Genetics October 18 22 2014 San Diego CA 15
20. y warranty or guaranteed support whatsoever Harvard University cannot be responsible for its use misuse or functionality The software may be freely copied for non commercial purposes provided this copyright notice is re tained Starting from v2 0 the BOLT REML component of the software also uses routines from the NLopt library written by Steven G Johnson and distributed under the MIT License Copy right 2007 2011 Massachusetts Institute of Technology 13 References 1 10 11 12 13 14 15 Loh P R et al Efficient Bayesian mixed model analysis increases association power in large cohorts Nature Genetics 2015 Kang H M et al Variance component model to account for sample structure in genome wide association studies Nature Genetics 42 348 354 2010 Lippert C et al FaST linear mixed models for genome wide association studies Nature Methods 8 833 835 2011 Listgarten J et al Improved linear mixed models for genome wide association studies Nature Methods 9 525 326 2012 Listgarten J Lippert C amp Heckerman D FaST LMM Select for addressing confounding from spatial structure and rare variants Nature Genetics 45 470 471 2013 Lippert C et al The benefits of selecting phenotype specific variants for applications of mixed models in genomics Scientific Reports 3 2013 Zhou X amp Stephens M Genome wide efficient mixed model analysis for as
21. ysical coordinates of your bim file You may use the geneticMapFile option even if your PLINK bim file does contain genetic coordinates in this case the genetic coordinates in the bim file will be ignored and interpolated coordinates will be used instead 5 1 2 Imputed SNP dosages As of version 1 1 the BOLT LMM association testing algorithm supports computation of mixed model association statistics at an arbitrary number of imputed SNPs with real valued dosages rather than hard called genotypes using a mixed model built on a subset of hard called genotypes BOLT REML variance components analysis does not support dosage input This approach re quires only a trivial amount of additional computation and no additional RAM as it simply applies 6 a genome scan as in GRAMMAR Gamma 8 of real valued dosage SNPs against the residual phenotypes that BOLT LMM already computes We expect that using a mixed model built on only a subset of 500K hard called genotypes should sacrifice almost no power while retaining the computational efficiency of BOLT LMM If you have only imputed SNP data on hand you will need to pre process your data set to create a subset of hard called SNPs in PLINK format for BOLT LMM We suggest the following procedure 1 Determine a high confidence set of SNPs e g based on IMPUTE2 INFO score at which to create an initial hard call set 2 Create hard called genotypes at these SNPs in PLINK format 3 Use
Download Pdf Manuals
Related Search