There's at least one, possibly two problems here. I'm trying to use MVP.Data to prepare my data in PLINK binary format for FarmCPU. I first noticed something was possibly wrong when MVP had written about 25% of the SNP markers (8.2M total)... at this point, the .geno.bin output stopped increasing in size. It had reached 7GB and stayed at this size from the time the console said "Number of markers written into BIG file" was ~2M to when it was ~8M. The .bed input is 28GB, yet the largest output was this 7GB .geno.bin.

Also, I appear to be unable to make a kinship matrix once writing markers into the BIG file is complete. This problem is shown below, below the command I used.

>MVP.Data(fileBed="my_in_prefix",
+          filePhe="my_pheno_data",
+          fileKin=TRUE,
+          filePC=TRUE,
+          out="my_out_prefix"    
+ )

[1] "Number of Markers Written into BIG File: 8252993"
[1] "Preparation for Genotype data is done!"
[1] "Preparation for PHENOTYPE data is done!"
[1] "Calculate KINSHIP using Vanraden method..."
Error: cannot allocate vector of size 56.4 Gb

This was followed by R crashing in the way it normally does when too much memory is used. I thought 64GB RAM would be enough for our 28GB SNP set, so that's all I asked for. Our admins can quickly add more to the virtual machine I'm working on if needed. If this is the problem, how much RAM should they add for us to be sure the whole process (from MVP.Data through FarmCPU) will work?

I have a kinship matrix prepared by EMMAX-kin, so would it be more wise for me to import and parse that one than to try making a new kinship matrix?

At this point in time, we have to use this large SNP set, which has already been filtered based on 5% maf... We can do more filtering later, but first need to make sure we can replicate results with the 8.2M set across multiple platforms.

Thanks much for your time!

Hi,

I am new to GitHub, so please excuse me if my question is not issued in a proper way.

I have a question regarding the permutation threshold. If I understand it correctly, the 95% quantile of the p-values vector is used as a significance threshold and I found that to be relatively high or maybe not stringent enough (see below an example with 200 permutations).

The authors of this paper (https://onlinelibrary.wiley.com/doi/full/10.1111/tpj.14659) used a 95% quantile not of the p-value itself but the -log10(p) value.

Figure 5: Arouisse, B., Korte, A., van Eeuwijk, F., & Kruijer, W. (2020). Imputation of 3 million SNPs in the Arabidopsis regional mapping population. The Plant Journal, 102(4), 872–882. https://doi.org/10.1111/tpj.14659
(this is a different trait as the one I tested, but the experiment is in the same species, with a similar amount of genotypes as I used and I obtained the markers from their imputation)

Following that approach I was wondering whether this would correspond to a 5% quantile if the p-value itself is used. Maybe this is a matter of preference, but I was thinking that having a relatively stringent p-value, while still detecting significant results would be a good trade-off between type I and II errors. Maybe I am understanding the math/statistics behind it incorrectly, but I would appreciate it if you could give a short comment.
Thanks a lot in advance!

Hi. How to set parameter for showing all traits in separated axes ?

Hi,
Thanks for developing this nice package.
I want to know why am I getting NA value for se in some significant SNPs? also, is there a way to calculate proportion of variance explained by the significant SNP?

Thanks for your time!

hc.FarmCPU_signals.txt

Hi,

I'm struggling to get my numeric data to read in MVP.

I prepared the numeric format based on the instructions at https://github.com/XiaoleiLiuBio/MVP

It looks like this (Just a portion shown here. The total number of row/SNP is 52157 and total number of column/individual is 68), no header and row names:

NaN | NaN | NaN | NaN | NaN | NaN | NaN
0 | 0 | 0 | 0 | 2 | 0 | NaN
2 | 2 | 2 | 2 | 2 | 2 | 0
2 | 2 | 2 | 2 | 2 | 2 | 2
2 | 0 | 2 | 2 | 2 | 0 | 0
0 | NaN | 0 | 0 | 0 | 2 | 2
NaN | NaN | NaN | NaN | NaN | NaN | NaN
0 | 0 | 0 | 0 | 0 | 0 | 0

I understand that there shouldn't be NaN or missing data in numeric format. I don't know how to overcome this. Could you please help me?

I also got this error:

Error in big.matrix(nrow = numRows, ncol = createCols, type = type, dimnames = list(rowNames, :
A big.matrix must have at least one row and one column

I used this code to try to fix:

fileNum=matrix(1:3546676,ncol=68)

But still got the same error.

May I ask what should I do?

Appreciate your help.

Many thanks.

Regards,
Tina

Hi Xiaolei,
I got geno, pheno data and map ready with MVP.Data function for the run, but when I start the run, I got the following error:

[1] "Input data has 198 individuals, 109914 markers"
[1] "Principal Component Analysis Start ..."
means for first 10 snps:
[1] 0 0 0 0 0 0 0 0 0 0
Error in SetAll.bm(x, value) :
Matrix dimensions do not agree with big.matrix instance set size.

Any idea why this error occurred? Thanks!

where the PVE value by each SNP?

thank you!

Hi,

I am trying to install MVP in my Mac, but when I typed > source("MVPinstall.r") I get this error message. What does this mean? and what should I do?

> Error in install.packages(pkg, repos = NULL) : 
>   type == "both" cannot be used with 'repos = NULL'

My R versions is

$version.string
[1] "R version 3.3.3 (2017-03-06)"

Thanks in advance,

Regards,

Ángela Parody-Merino

Hi Xiaolei,
I am wondering what's the best way to cite MVP in my manuscript. Thanks!

Hi

Thank you for the recent improvements to this package!

I have two minor things:

1. When rMVP reports the significance level, it cuts off at 6 decimal places:

   rMVP/R/MVP.r

   Line 307 in 3acd666

   logging.log(paste("Significant level: ", sprintf("%.6f", threshold/m), sep=""), "\n", verbose = verbose)

   
   In my case I have so many SNPs that rMVP always prints Significant level: 0.000000. Perhaps it would be better to use scientific notation using formatC(0.0000000000005, format = "e", digits = 2)?

2. The files ending in signal.csv are very useful, is it possible there is a bug when only a single SNP is significant (which happens a lot with FarmCPU)?
   Normally, I get a proper table with proper headers, but when I have only one SNP, I get a strange text-file like this:

"x"
"SNP_5015"
"3"
"47924314"
"-6185.31927432418"
"1015.52184827578"
"1.50018111376143e-08"

i.e., the single-row table got transposed. Funny enough it's just a cbind and a write.csv, I currently don't understand how this happens.

sessionInfo():

R version 3.6.1 (2019-07-05)
Platform: x86_64-conda_cos6-linux-gnu (64-bit)
Running under: Red Hat Enterprise Linux Server release 6.10 (Santiago)

locale:
 [1] LC_CTYPE=en_AU.UTF-8       LC_NUMERIC=C
 [3] LC_TIME=en_AU.UTF-8        LC_COLLATE=en_AU.UTF-8
 [5] LC_MONETARY=en_AU.UTF-8    LC_MESSAGES=en_AU.UTF-8
 [7] LC_PAPER=en_AU.UTF-8       LC_NAME=C
 [9] LC_ADDRESS=C               LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_AU.UTF-8 LC_IDENTIFICATION=C

attached base packages:
[1] parallel  stats     graphics  grDevices utils     datasets  methods
[8] base

other attached packages:
[1] rMVP_0.99.16     bigmemory_4.5.33 MASS_7.3-51.4

loaded via a namespace (and not attached):
[1] bigmemory.sri_0.1.3 compiler_3.6.1      tools_3.6.1
[4] Rcpp_1.0.2

There are notes in the build report that seem like true positives, and that should be addressed:

• seem like straight-forward corrections

  * checking R code for possible problems ... [16s/16s] NOTE
  MVP.Data.Kin: warning in read.big.matrix(fileKin, head = FALSE, type =
    "double", sep = sep): partial argument match of 'head' to 'header'
  MVP.Data.Numeric2MVP: warning in read.big.matrix(num_file, head =
    FALSE, sep = sep): partial argument match of 'head' to 'header'
  MVP.Data.PC: warning in read.big.matrix(filePC, head = FALSE, type =
    "double", sep = sep): partial argument match of 'head' to 'header'
  MVP.FarmCPU: warning in FarmCPU.BIN(Y = phe[, c(1, 2)], GD = geno, GM =
    map, CV = CV, P = myPrior, method = method.bin, b = bin.size, s =
    bin.selection, theLoop = theLoop, bound = bound, ncpus = ncpus):
    partial argument match of 'GD' to 'GDP'
  MVP.FarmCPU: warning in FarmCPU.BIN(Y = phe[, c(1, 2)], GD = geno, GM =
    map, CV = theCV, P = myPrior, method = method.bin, b = bin.size, s =
    bin.selection, theLoop = theLoop, ncpus = ncpus): partial argument
    match of 'GD' to 'GDP'
  MVP.FarmCPU: warning in FarmCPU.Remove(GD = geno, GM = map, seqQTN =
    seqQTN, seqQTN.p = seqQTN.p, threshold = 0.7): partial argument match
    of 'GD' to 'GDP'
  print_info: warning in make_line(welcome, line = linechar, width =
    width): partial argument match of 'line' to 'linechar'
  

• Undefined global functions or variables: Revo.version getMKLthreads setMKLthreads looks like missing imports
• seems like mis-use of compiler flags

  * checking use of SHLIB_OPENMP_*FLAGS in Makefiles ... NOTE
    src/Makevars: SHLIB_OPENMP_CXXFLAGS is included in PKG_CXXFLAGS but not in PKG_LIBS
    src/Makevars: SHLIB_OPENMP_CFLAGS is included in PKG_LIBS but linking is by C++
    src/Makevars.win: SHLIB_OPENMP_CXXFLAGS is included in PKG_CXXFLAGS but not in PKG_LIBS
    src/Makevars.win: SHLIB_OPENMP_CFLAGS is included in PKG_LIBS but linking is by C++
  Use of these macros is discussed in sect 1.2.1.1 of Writing R
  Extensions. The macros for different languages may differ so the
  matching macro must be used in PKG_CXXFLAGS (etc) and match that used
  in PKG_LIBS (except for Fortran: see the manual).
  

DESCRIPTION

• Please provide a Description: field that reads more like a paper abstract, providing the user with an informative summary of the functionality of your package and how it advances current 'state of the art'. Remember that the field can be wrapped to multiple lines, with continuation lines indented, e.g., by 4 spaces.
• It seems likely that BH, RccpPrgoess, Rcpp, parallel and perahps other packages are not 'seen' by the end user, and should be in Imports: rather than Depends:.

NAMESPACE

• BH does not have any R functions, so importing it's NAMESPACE does not make sense

vignettes

• Please arrange for more of your vignette code chunks to be actually evaluated. The problem is that these 'pictures of code' quickly break, and the user is left with the impression that the package functions as advertised when it does not.

R -- the following are specific comments, but issues should be addressed throughout the package

• MVP.Data.r:77 message("WARNING...") if this is a warning, then use warning() rather than than message().
• MVP.Data.r:77 please avoid long lines that combine programming idioms, they are difficult to read and comprehend

  if ("sep.vcf" %in% names(parames))
      warning("'sep.vcf' has be DEPRECATED")
  

• MVP.Data.r:90 use vapply() rather than sapply() to avoid surprises when the length of the first argument is 0.
• MVP.Data.r:223 use message() rather than cat() to communicate informative messages to users; note that message() does not usually require paste().
• MVP.Data.r:550 use seq_len() or seq_along() rather than 1:n to avoid surprises when n < 1.
• MVP.Data.r:555 stop("ERRROR: ...") and similar seems redundant, simply stop("...") would seem sufficient.
• MVP.Data.r:857 setting options(...) without restoring values to their default is bad style. use

  opts <- options(...)
  on.exit(options(opts))
  

• MVP.EMMA.Vg.Ve.r:113 another example of using seq_len() rather than 1:(m - 1)
• MVP.EMMA.Vg.Ve.r: 115 'hoist' loop invariant code like this function definition outside the for loop to avoid redundant computation.
• MVP.FarmCPU.r:95 another example where use of message() is appropriate
• MVP.FarmCPU.r:188 reconsider the need for verbose output like this.
• MVP.FarmCPU.r:300 this code is highly similar (identical?) to other code; write simple functions and reuse rather than copy / paste.
• MVP.GLM.r:93 please use consistent indentation for readability
• MVP.PCA.r:230 remove dead code.
• MVP.Utility.r:59 re-think the value of re-implementing a progress bar; this represents additional code that you must maintain, providing limited incremental value.

General

• are the files on the master branch of this repository necessary for the package? For instance, are the files in the results folder used?

Unable execute the code getting error as
Error in MVP.Data(fileHMP = "HapMapwildsoy.hmp.txt", filePhe = "Pheno_269.txt", : could not find function "MVP.Data"

Hi! I get this error:

Error in if (Max <= 1) { : missing value where TRUE/FALSE needed
Calls: MVP.Report
Execution halted

library(rMVP)
MVP.Data(fileHMP="Filtered genotype.hmp copy.txt",
         filePhe="Phenotypic_data.txt",
         sep.hmp="\t",
         sep.phe="\t",
         SNP.effect="Add",
         fileKin=FALSE,
         filePC=FALSE,
         out="mvp.hmp",
         #priority="memory",
         #maxLine=10000
)

genotype <- attach.big.matrix("mvp.hmp.geno.desc")
phenotype <- read.table("mvp.hmp.phe",head=TRUE)
map <- read.table("mvp.hmp.geno.map" , head = TRUE)

imMVP <- MVP(
  phe=phenotype,
  geno=genotype,
  map=map,
  #K=Kinship,
  #CV.GLM=Covariates,     ##if you have additional covariates, please keep there open.
  #CV.MLM=Covariates,
  #CV.FarmCPU=Covariates,
  nPC.GLM=5,      ##if you have added PC into covariates, please keep there closed.
  nPC.MLM=3,
  nPC.FarmCPU=3,
  priority="speed",
  #ncpus=10,
  vc.method="BRENT",
  maxLoop=10,
  method.bin="FaST-LMM",#"FaST-LMM","EMMA", "static"
  #permutation.threshold=TRUE,
  #permutation.rep=100,
  threshold=0.05,
  method=c("GLM", "MLM", "FarmCPU")
)

This is the output that I am getting

> library(rMVP)
> MVP.Data(fileHMP="Filtered genotype.hmp copy.txt",
+          filePhe="Phenotypic_data.txt",
+          sep.hmp="\t",
+          sep.phe="\t",
+          SNP.effect="Add",
+          fileKin=FALSE,
+          filePC=FALSE,
+          out="mvp.hmp",
+          #priority="memory",
+          #maxLine=10000
+ )
Preparing data for MVP...
Reading file...
inds: 133	markers:1472
Number of threads: 1
0%   10   20   30   40   50   60   70   80   90   100%
[----|----|----|----|----|----|----|----|----|----|
**************************************************|
Preparation for GENOTYPE data is done within 0s 
131 common individuals between phenotype and genotype. 
Preparation for PHENOTYPE data is Done within 0s 
Imputing...
out is NULL, impute inplace.
Number of threads: 3
0%   10   20   30   40   50   60   70   80   90   100%
[----|----|----|----|----|----|----|----|----|----|
**************************************************|
Impute Genotype File is done!
MVP data prepration accomplished successfully!
Warning messages:
1: In MVP.Data(fileHMP = "Filtered genotype.hmp copy.txt", filePhe = "Phenotypic_data.txt",  :
  SNP.effect has been DEPRECATED.
2: In MVP.Data(fileHMP = "Filtered genotype.hmp copy.txt", filePhe = "Phenotypic_data.txt",  :
  sep.hmp has been DEPRECATED.
> genotype <- attach.big.matrix("mvp.hmp.geno.desc")
> phenotype <- read.table("mvp.hmp.phe",head=TRUE)
> map <- read.table("mvp.hmp.geno.map" , head = TRUE)
> imMVP <- MVP(
+   phe=phenotype,
+   geno=genotype,
+   map=map,
+   #K=Kinship,
+   #CV.GLM=Covariates,     ##if you have additional covariates, please keep there open.
+   #CV.MLM=Covariates,
+   #CV.FarmCPU=Covariates,
+   nPC.GLM=5,      ##if you have added PC into covariates, please keep there closed.
+   nPC.MLM=3,
+   nPC.FarmCPU=3,
+   priority="speed",
+   #ncpus=10,
+   vc.method="BRENT",
+   maxLoop=10,
+   method.bin="FaST-LMM",#"FaST-LMM","EMMA", "static"
+   #permutation.threshold=TRUE,
+   #permutation.rep=100,
+   threshold=0.05,
+   method=c("GLM", "MLM", "FarmCPU")
+ )
====================== Welcome to MVP ======================
      A Memory-efficient, Visualization-enhanced, and       
             Parallel-accelerated Tool For GWAS             
                    __  __  __   __  ___                    
                   |  \/  | \ \ / / | _ \                   
                   | |\/| |  \ V /  |  _/                   
                   |_|  |_|   \_/   |_|   Version: 0.99.17  
  Designed and Maintained by Lilin Yin, Haohao Zhang, and   
  Xiaolei Liu                                               
  Contributors: Zhenshuang Tang, Jingya Xu, Dong Yin,       
  Zhiwu Zhang, Xiaohui Yuan, Mengjin Zhu, Shuhong Zhao,     
  Xinyun Li                                                 
  Contact: [email protected]                      
============================================================
Start: 2020-03-03 15:19:31 
The log has been output to the file: /Users/rudrakshisharma/Desktop/Manhattan plots/MVP.20200303_151931.log 
Input data has 131 individuals, 1472 markers 
Phenotype: Arimont 
Relationship matrix mode in speed 
Scale the genotype matrix 
Computing Z'Z 
Deriving relationship matrix successfully 
Eigen Decomposition 
Deriving PCs successfully 
Number of PCs included in GLM: 5 
Number of PCs included in MLM: 3 
Number of PCs included in FarmCPU: 3 
-------------------------GWAS Start------------------------- 
General Linear Model (GLM) Start... 
scanning...

Mixed Linear Model (MLM) Start... 
Variance components using: BRENT 
Estimated Vg and Ve: NA NA 
Error in svd(X) : infinite or missing values in 'x'
In addition: Warning messages:
1: In MVP.GLM(phe = phe, geno = geno, CV = CV.GLM, cpu = ncpus, bar = bar,  :
  NAs introduced by coercion
2: In MVP.MLM(phe = phe, geno = geno, K = K, eigenK = eigenK, CV = CV.MLM,  :
  NAs introduced by coercion

Can someone please help me on this issue??

Regards
Rudra

Hi,
I recently got this incomplete manhatten plot while running a rMVP script on a linux server (end of Chr 12 not mapped) This phenomenon does not occur when I am running the same script on my windows working station.
Thanks in advance for the help.

install.packages("rMVP")
Installing package into ‘C:/Users/user/Documents/R/win-library/3.5’
(as ‘lib’ is unspecified)
package ‘rMVP’ is not available (for R version 3.5.3)

Or the the development version from GitHub:

install.packages("devtools")

devtools::install_github("xiaolei-lab/rMVP")
Error in loadNamespace(j <- i[[1L]], c(lib.loc, .libPaths()), versionCheck = vI[[j]]) :
there is no package called ‘glue’

Or the the development version from GitHub:

install.packages("devtools")

devtools::install_github("xiaolei-lab/rMVP")
Error in loadNamespace(j <- i[[1L]], c(lib.loc, .libPaths()), versionCheck = vI[[j]]) :
there is no package called ‘glue’
devtools::install_github("xiaolei-lab/rMVP")
Error in loadNamespace(j <- i[[1L]], c(lib.loc, .libPaths()), versionCheck = vI[[j]]) :
there is no package called ‘glue’

How to choose values of nPC.GLM, nPC.MLM and nPC.FarmCPU?
And, How to choose values of nPC.GLM, nPC.MLM and nPC.FarmCPU when I use K, CV.GLM, CV.MLM, CV.FarmCPU .

imMVP <- MVP(
    phe=phenotype,Ff
    geno=genotype,
    map=map,
    #K=Kinship,
    #CV.GLM=Covariates,
    #CV.MLM=Covariates,
    #CV.FarmCPU=Covariates,
    nPC.GLM=5,
    nPC.MLM=3,
    nPC.FarmCPU=3,
    priority="speed",
    #ncpus=10,
    vc.method="BRENT",
    maxLoop=10,
    method.bin="FaST-LMM",#"FaST-LMM","EMMA", "static"
    #permutation.threshold=TRUE,
    #permutation.rep=100,
    threshold=0.05,
    method=c("GLM", "MLM", "FarmCPU")
)

Hello, Xiaolei,

I found single trait perform correctly, multiple traits analysis running with errors.(version 0.99.10)

for(i in 2:ncol(phenotype)){imMVP <- MVP(phe=phenotype[, c(1, i)],geno=genotype,map=map,K=Kinship,nPC.GLM=5,nPC.MLM=3,nPC.FarmCPU=3,perc=1,priority="speed",vc.method="GEMMA",maxLoop=10,method.bin="FaST-LMM",permutation.threshold=TRUE,permutation.rep=100.threshold=0.05,method=c("GLM", "MLM", "FarmCPU"))}
Error: unexpected symbol in " 2:ncol(phenotype)){imMVP <- MVP(phe=phenotype[, c(1, i)],geno=genotype,map=map,K=Kinship,nPC.GLM=5,nPC.MLM=3,nPC.FarmCPU=3,perc=1,priority="speed",vc.method="GEMMA",maxLoop=10,method.bin="FaS"
Execution halted

Hi,

I have 2 questions:

Q1. I meet this problem at the end when running MVP:-

Error in big.PCA(bigMat = big.geno, pcs.to.keep = pcs.keep) :
could not find function "big.PCA"

Here is my script:

> imMVP <- MVP(
+     phe=phenotype,
+     geno=genotype,
+     map=map,
+     #K=Kinship,
+     #CV.GLM=Covariates,
+     #CV.MLM=Covariates,
+     #CV.FarmCPU=Covariates,
+     nPC.GLM=3,
+     nPC.MLM=3,
+     nPC.FarmCPU=3,
+     perc=1,
+     priority="speed",
+     ncpus=12,
+     vc.method="EMMA",
+     maxLoop=10,
+     method.bin="FaST-LMM",#"FaST-LMM","EMMA", "static"
+     #permutation.threshold=TRUE,
+     #permutation.rep=100,
+     threshold=0.05,
+     method=c("GLM", "MLM", "FarmCPU")
+ )
#--------------------------------------------Welcome to MVP---------------------------------------------# 
#          A Memory-efficient, Visualization-enhanced, and Parallel-accelerated Tool For GWAS           # 
# Version: 1.0.1                                                                                        # 
# Authors: Lilin Yin, Haohao Zhang, Zhiwu Zhang, Xinyun Li, Xiaohui Yuan, Shuhong Zhao, and Xiaolei Liu # 
# Contact: [email protected]                                                                  # 
#-------------------------------------------------------------------------------------------------------# 
[1] "Input data has 34 individuals, 51910 markers"
[1] "Principal Component Analysis Start..."
Error in big.PCA(bigMat = big.geno, pcs.to.keep = pcs.keep) : 
  could not find function "big.PCA"

Q2. I have 38 individuals but it appears that it is reading only 34?

My MVP.Data script looks like this and was successful:

> MVP.Data(fileHMP="~/Desktop/SNP_formatted.hmp.txt",
+          filePhe="~/Desktop/Phenotype.txt",
+          sep.hmp="\t",
+          sep.phe="\t",
+          SNP.effect="Add",
+          fileKin=FALSE,
+          filePC=FALSE,
+          out="mvp.hmp",
+          #priority="memory",
+          #maxLine=10000
+ )
[1] "Preparing data for MVP..."
[1] "Preparation for PHENOTYPE data is done!"
[1] "Output numeric genotype..."
[1] "File: ~/Desktop/myGAPIT/SNP_Rlm7_formatted.hmp.txt ; Total markers:  51910  finished!"
[1] "Preparation for numeric data is done!"
[1] "Output mvp genotype..."
[1] "MVP data prepration accomplished successfully!"

I added my data using the MVP.data function. I've confirmed that the output files all have the same n value, but when I run the following code, I get a Matrix dimensions error. This error does not occur if I run the model without PCs.
I noticed in the MVP readme, that the Covariate lines were commented out. Is this a problem with the underlying packages?

MVP.Data(fileHMP="data/filtered.hmp.txt",
         filePhe="data/mvp_input_20180511.txt",
         sep.hmp="\t",
         sep.phe="\t",
         SNP.effect="Add",
         fileKin=T,
         filePC=T,
         out="mvp"
)

genotype <- attach.big.matrix("mvp.geno.desc")
phenotype <- read.table("mvp.phe",head=TRUE)
map <- read.table("mvp.map" , head = TRUE)
Kinship <- attach.big.matrix("mvp.kin.desc")
Covariates <- attach.big.matrix("mvp.pc.desc")

for(i in 2:ncol(phenotype)){
  chiles_env_pcs <- MVP(
    phe=phenotype[, c(1, i)],
    geno=genotype,
    map=map,
    K=Kinship,
    CV.GLM=Covariates,
    CV.MLM=Covariates,
    CV.FarmCPU=Covariates,
    nPC.GLM=5,
    nPC.MLM=3,
    nPC.FarmCPU=3,
    perc=1,
    priority="memory",
    ncpus=2,
    vc.method="GEMMA",
    maxLoop=10,
    method.bin="FaST_LMM",#"FaST-LMM","EMMA", "static"
    threshold = 1.895353e-06,
    method=c("GLM", "MLM", "FarmCPU")  )
}

The Error Message:

[1] "Input data has 190 individuals, 30418 markers"
[1] "Principal Component Analysis Start..."
 means for first 10 snps:
 [1] 0 0 0 0 0 0 0 0 0 0
Error in SetElements.bm(x, i, j, value) : 
  Matrix dimensions do not agree with big.matrix instance set size.

Hi,

I ran FarmCPU for multiple traits and I got quite unusual results. I attach a few manhattan plots to illustrate the issue.

Cheers,

Stephan

Hi,

I'm calculated the number of effective SNPs and was wondering how I would be able to change marker size to this value so that I can have a corrected Bonferroni threshold according to the number of effective SNPs. How do I change that in rMVP? I tried doing threshold = 0.05/effective size but that seems to inflate the threshold value even more. Any suggestions would be appreciated!

@XiaoleiLiuBio

Dear professor Liu,

Thank you for providing this wonderful tool.

In practice, many gwas summary based downstream analyses required infomation like "Number of tested samples", "Allele1 (the minor allele)" and "Allele2 (the major allele)". Although these data could be extracted from the input, it is convenient that the MVP could output these data by default.

Best wishes,

Songtao Gui

I tried the two methods you write on GitHub but both of them failed.

devtools::install_github("xiaoleiLiubio/[email protected]")
Downloading GitHub repo xiaoleiLiubio/[email protected]
Error: Failed to install 'rMVP' from GitHub:
Could not find tools necessary to compile a package
Call pkgbuild::check_build_tools(debug = TRUE) to diagnose the problem.

install.packages("rMVP_0.99.14.1.tar.gz", repos=NULL)
Warning: invalid package ‘rMVP_0.99.14.1.tar.gz’
Error: ERROR: no packages specified

How to change grey color to white color?

The issue occurs when trying to run MVP(). It is as follows:

genotype = attach.big.matrix("/work/schnablelab/jhurst5/mvp2.hmp.numeric")
Error in parse(file = file, keep.source = keep.source) :
mvp2.hmp.numeric:1:9: unexpected numeric constant
1: 0   0
            ^
Calls: attach.big.matrix ... tryCatchOne -> -> dget -> eval -> parse
Execution halted

Nothing seems wrong with the column or row numbers in the data.

If anyone knows what the problem is, your help would be appreciated.

dear liu:

when I was run a MVP, I figure that you segess code is

> for(i in 2:ncol(phenotype)){
+   imMVP <- MVP(
+     phe=phenotype[, c(1, i)],
+     geno=genotype,
+     map=map,
+     nPC.GLM=5,
+     nPC.MLM=3,
+     nPC.FarmCPU=3,
+     perc=1,
+     priority="memory",
+     ncpus=16,
+     vc.method="EMMA",
+     maxLoop=10,
+     method.bin="FaST-LMM",
+     threshold=0.05,
+     method=c("GLM", "MLM", "FarmCPU"),
+ file="pdf",
+   )
+ }

can I just Run

> imMVP <- MVP(
+     phe=phenotype,
+     geno=genotype,
+     map=map,
+     nPC.GLM=5,
+     nPC.MLM=3,
+     nPC.FarmCPU=3,
+     perc=1,
+     priority="memory",
+     ncpus=16,
+     vc.method="EMMA",
+     maxLoop=10,
+     method.bin="FaST-LMM",
+     threshold=0.05,
+     method=c("GLM", "MLM", "FarmCPU"),
+ file="pdf",
+   )

Hi Xiaolei,
thank you for this fantastic tool. I am using permutations to determine a significance threshold for a FarmCPU GWA. I see that the permuted threshold is correctly plotted to the resulting manhattan plots, however I cannot find the numerical value of the threshold saved anywhere in the object returned by MVP() function. Am I missing something? Thank you so much

Matteo

rfunctions install error!

I install rfunctions
Spectra/LinAlg/TridiagEigen.h:93:49: error: no matching function for call to 'tridiagonal_qr_step(double*&, double*&, int&, int&, Eigen::PlainObjectBase<Eigen::Matrix<double, -1, -1> >::Scalar*, int&)'
Eigen::internal::tridiagonal_qr_step(maind, subd, start, end, evecs.data(), n);
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from /mnt/ilustre/users/dna/.env/lib64/R/library/RcppEigen/include/Eigen/Eigenvalues:39:0,

Hi Xiaolei,

first - thank you very much for information about your new package. It looks very nice and promising, i am under impressed.

I would like to know is there any way to work with MVP with missing value in phenotype? I generate MVP data (phenotype file have some missing data) and i try to start analysis (only FarmCPU) but i got following error:
when i used FaST-LMM option:

Error in if (llresults[[i]]$LL > LL.save) { :
missing value where TRUE/FALSE needed
In addition: Warning messages:
1: In FUN(newX[, i], ...) : no non-missing arguments to min; returning Inf
2: In FUN(newX[, i], ...) : no non-missing arguments to min; returning Inf
3: In min(P[P != 0], na.rm = TRUE) :
no non-missing arguments to min; returning Inf
4: In if (!is.na(inclosure.size)) { :
the condition has length > 1 and only the first element will be used

and when i use static:

Error in seq.int(from, to, length.out = n) : 'from' must be finite
In addition: Warning messages:
1: In FUN(newX[, i], ...) : no non-missing arguments to min; returning Inf
2: In FUN(newX[, i], ...) : no non-missing arguments to min; returning Inf
3: In min(P[P != 0], na.rm = TRUE) :
no non-missing arguments to min; returning Inf
4: In if (!is.na(inclosure.size)) { :
the condition has length > 1 and only the first element will be used
5: In min(myPrior, na.rm = TRUE) :
no non-missing arguments to min; returning Inf
6: In min(P[P != 0], na.rm = TRUE) :
no non-missing arguments to min; returning Inf

You can repeat this error by setting up any value in demo data with NA. In basic FarmCPU package my data works well.

Best,
Marcin

Hi Xiaolei,
thanks a lot for writing this nice package.
I would like to use MVP for my project, but I am not sure the best way to convert unphased SNP genotype (vcf format, around 60k SNPs) to numeric matrix (0,1,2). Do you which tool I should use to do the conversion.

Hi,

does indel coded as "--" are intentionally not supported? I have HapMap file, which work properly with TASSEL, however after running MVP.Data() on it, indels are coded same as alternative allele. For exaple line:

[1] "S01_17126" "A/-" "01" "17126" "+" NA NA NA NA NA NA "--" [13] "AA" "--" "AA" "--" "AA" "--" "AA" "AA" "--" "--" "AA" "AA" [25] "--" "AA" "AA" "AA" "AA" "--" "AA" "AA" "AA" "AA" "AA" "AA" [37] "--" "AA" "--" "--" "AA" "--" "--" "--" "--" "--" "--" "AA" [49] "--" "--" "AA" "--" "AA" "AA" "AA" "--" "AA" "AA" "AA" "AA" [61] "AA" "--" "--" "--" "AA" "--" "AA" "--" "AA" "--" "AA" "AA" [73] "--" "--" "--" "--" "--" "AA" "AA" "--" "--" "--" "--" "--" [85] "--" "--" "--" "--" "AA" "--" "--" "AA" "--" "--" "AA" "AA" [97] "--" "--" "AA" "AA" "AA" "--" "--" "--" "AA" "AA" "AA" "AA" [109] "AA" "AA" "AA" "AA" "AA" "--" "AA" "--" "AA" "--" "AA" "AA" [121] "--" "AA" "AA" "AA" "AA" "AA" "--" "AA" "AA" "AA" "--" "AA" [133] "AA" "--" "AA" "AA" "AA" "AA" "AA" "AA" "AA" "AA" "--" "--" [145] "--" "AA" "AA" "AA" "AA" "AA" "--" "AA" "AA" "--" "AA" "AA" [157] "--" "--" "--" "AA" "--" "--" "AA" "AA" "AA" "--" "AA" "AA" [169] "AA" "--" "--" "--" "AA" "AA" "AA" "AA" "AA" "--" "--" "AA" [181] "AA" "AA" "AA" "--" "--" "AA" "--" "AA" "--" "--" "AA" "AA" [193] "--" "AA" "AA" "AA" "--" "AA" "--" "AA" "AA" "AA" "AA" "--" [205] "AA" "--" "--" "--" "AA" "--" "AA" "--" "--" "--" "--" "AA" [217] "--" "A-" "--" "AA" "AA" "AA" "AA" "AA" "AA" "--" "--" "--" [229] "--" "AA" "--" "--" "AA" "--" "--" "--" "--" "AA" "--" "--" [241] "AA" "--" "AA" "AA" "AA" "AA" "--" "AA" "--" "--" "--" "-A" [253] "AA" "AA" "-A" "--" "AA" "--" "AA" "AA" "AA" "--" "AA" "AA" [265] "AA" "AA" "--" "AA" "--" "AA" "AA" "AA" "--" "--" "--" "AA" [277] "--" "AA" "AA" "--" "--" "--" "AA" "--" "--" "--" "--" "--" [289] "AA" "AA" "AA" "AA" "AA" "AA" "--" "AA" "AA" "--" "AA" "AA" [301] "AA" "--" "AA" "AA" "--" "AA" "AA" "AA" "--" "--" "AA" "--" [313] "AA" "AA" "--" "AA" "--" "AA" "--" "--" "AA" "AA" "AA" "AA" [325] "--" "--" "AA" "--" "--" "--" "--" "--" "AA" "--" "--" "AA" [337] "--" "--" "--" "--" "--" "--" "AA" "--" "AA" "AA" "--" "--" [349] "AA" "--" "AA" "--" "AA" "AA" "--" "AA" "AA" "AA" "AA" "AA" [361] "--" "AA" "AA" "--" "--" "AA" "--" "AA" "AA"

in big.matrix object created with MVP.Data looks like this:

[1] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 [76] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 [151] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 [226] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 [301] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

In map file alleles are properly coded:
S01_17126 1 17126 A -

Rest SNPs looks fine. I have used most recent version rMVP installed from github.

Best,
Marcin

There is a R package "export" which can export figure to ppt.
So can you add this function to MVP.Report?

library(export)
library(ggplot2)
data <- read.table('~/Desktop/genetic_map/summary_genetic_physical_relationship.txt',sep='\t',header=T)
ggplot(data,aes(x=Physcial,y=Genetic)) + geom_point(size=0.4) 
graph2ppt( file="~/Desktop/genetic_map/genetic_physical.pptx", paper="A4")

Hello,

I am currently using MVP.Report() to make a multi trait Manhattan plot. There are two issues I am running into:

1. Is there an argument to change the title of the plot?

2. Is there an argument to change/add colors to the overlapping point-plot. It appears that there are only 4 colors by default(blue, green, purple and orange) so when a 5th phenotype is added to the plot it goes back to blue, making two different phenotypes have the same color for their respective points.

Any help would be greatly appreciated.

dear Liu，I got a problem in using ‘FarmCPU’ model.
MVP(phe=pheno,geno=geno,map=map,method=c('FarmCPU'))
Input data has 159 individuals, 1565878 markers
|------------------------GWAS Start------------------------|
FarmCPU Start...
Current loop: 1 out of maximum of 10
seqQTN:
NULL
scanning...
number of covariates in current loop is:
0

|> 1.00%
|-> 2.00%
|-> 3.00%
|--> 4.00%
|--> 5.00%
|---> 6.00%
|---> 7.00%
|----> 8.00%
|----> 9.00%
|-----> 10.00%
|-----> 11.00%
|------> 12.00%
|------> 13.00%
|-------> 14.00%
|-------> 15.00%
|--------> 16.00%
|--------> 17.00%
|---------> 18.00%
|---------> 19.00%
|----------> 20.00%
|----------> 21.00%
|-----------> 22.00%
|-----------> 23.00%
|------------> 24.00%
|------------> 25.00%
|-------------> 26.00%
|-------------> 27.00%
|--------------> 28.00%
|--------------> 29.00%
|---------------> 30.00%
|---------------> 31.00%
|----------------> 32.00%
|----------------> 33.00%
|-----------------> 34.00%
|-----------------> 35.00%
|------------------> 36.00%
|------------------> 37.00%
|-------------------> 38.00%
|-------------------> 39.00%
|--------------------> 40.00%
|--------------------> 41.00%
|---------------------> 42.00%
|---------------------> 43.00%
|----------------------> 44.00%
|----------------------> 45.00%
|-----------------------> 46.00%
|-----------------------> 47.00%
|------------------------> 48.00%
|------------------------> 49.00%
|-------------------------> 50.00%
|-------------------------> 51.00%
|--------------------------> 52.00%
|--------------------------> 53.00%
|---------------------------> 54.00%
|---------------------------> 55.00%
|----------------------------> 56.00%
|----------------------------> 57.00%
|-----------------------------> 58.00%
|-----------------------------> 59.00%
|------------------------------> 60.00%
|------------------------------> 61.00%
|-------------------------------> 62.00%
|-------------------------------> 63.00%
|--------------------------------> 64.00%
|--------------------------------> 65.00%
|---------------------------------> 66.00%
|---------------------------------> 67.00%
|----------------------------------> 68.00%
|----------------------------------> 69.00%
|-----------------------------------> 70.00%
|-----------------------------------> 71.00%
|------------------------------------> 72.00%
|------------------------------------> 73.00%
|-------------------------------------> 74.00%
|-------------------------------------> 75.00%
|--------------------------------------> 76.00%
|--------------------------------------> 77.00%
|---------------------------------------> 78.00%
|---------------------------------------> 79.00%
|----------------------------------------> 80.00%
|----------------------------------------> 81.00%
|-----------------------------------------> 82.00%
|-----------------------------------------> 83.00%
|------------------------------------------> 84.00%
|------------------------------------------> 85.00%
|-------------------------------------------> 86.00%
|-------------------------------------------> 87.00%
|--------------------------------------------> 88.00%
|--------------------------------------------> 89.00%
|---------------------------------------------> 90.00%
|---------------------------------------------> 91.00%
|----------------------------------------------> 92.00%
|----------------------------------------------> 93.00%
|-----------------------------------------------> 94.00%
|-----------------------------------------------> 95.00%
|------------------------------------------------> 96.00%
|------------------------------------------------> 97.00%
|-------------------------------------------------> 98.00%
|-------------------------------------------------> 99.00%
|-------------------------------------------------->100.00%
Current loop: 2 out of maximum of 10
Optimizing Pseudo QTNs...
Error in big.matrix(nrow = length(rows), ncol = length(cols), type = type, :
A big.matrix must have at least one row and one column
此外: Warning messages:
1: In FarmCPU.Specify(GI = GM, GP = GP, bin.size = b, inclosure.size = s) :
强制改变过程中产生了NA
2: In ID.GP/bin.size : 长的对象长度不是短的对象长度的整倍数
3: In FarmCPU.Specify(GI = GM, GP = GP, bin.size = b, inclosure.size = s) :
强制改变过程中产生了NA
4: In FarmCPU.Specify(GI = GM, GP = GP, bin.size = b, inclosure.size = s) :
强制改变过程中产生了NA
5: In FarmCPU.Specify(GI = GM, GP = GP, bin.size = b, inclosure.size = s) :
强制改变过程中产生了NA
I'm a beginner and I don't konw how to deal with it, please help me.

Hi Xiaolei,

Im want to test this new GWAS script, however, I have a problem installing the rfunction. What it seems to be the culprit?

Thanks in advance.

Best, Charles

"Warning: running command 'make -f "Makevars.win" -f "C:/PROGRA1/R/R-341.1/etc/x64/Makeconf" -f "C:/PROGRA1/R/R-341.1/share/make/winshlib.mk" SHLIB_LDFLAGS='$(SHLIB_CXXLDFLAGS)' SHLIB_LD='$(SHLIB_CXXLD)' SHLIB="rfunctions.dll" WIN=64 TCLBIN=64 OBJECTS="RcppExports.o bigmemory.o cgls.o geninv.o gls.o lanczos.o linop.o lsmr.o multivar_norm.o pnorm.o testing.o utils.o"' had status 2
ERROR: compilation failed for package 'rfunctions'"

devtools::install_github("xiaoleiLiubio/[email protected]")
Downloading GitHub repo xiaoleiLiubio/[email protected]
Error: Failed to install 'rMVP' from GitHub:
Could not find tools necessary to compile a package
In addition: Warning messages:
1: In untar2(tarfile, files, list, exdir) :
skipping pax global extended headers
2: In untar2(tarfile, files, list, exdir) :
skipping pax global extended headers

Hi,

I am receiving the following error:

Error in density.default(phe[, i]) : 'x' contains missing values

There is no blank data, and all missing values are NA.

If anyone can help, it would be greatly appreciated.

I want to know if MVP need imputed data; can it used for nonimpued data

Hi Xiaolei,
thank you for develop such a convenient package. I have some questions:
1, I hve read your paper on plos genetics, I think FarmCPU can detect epstasis loci. But new loci ware detected when previous significant SNP added to CV (somebody call this SNP-fix, p value of new loci was just excess the threshold). is that needed to add significant SNP as CV for FarmCPU model?
2, when I use permutation to set threshold on farmcpu model, MVP display "GLM", is that right?
can you tell me? (my statistics backgroud was weak, you just need tell me what was right)
thank you.

Hi Xiaolei,

When I use this command for the FarmCPU GWAS(showed below)

imMVP <- MVP(phe=phenotype,geno=genotype,map=map,K=Kinship,CV.FarmCPU=Covariates,nPC.FarmCPU=3,perc=1,priority="speed",ncpus=8,vc.method="EMMA",maxLoop=10,method.bin="FaST-LMM",permutation.threshold=TRUE,permutation.rep=100,threshold=0.05,method="FarmCPU")


#--------------------------------------Welcome to MVP--------------------------------------# 
#    A Memory-efficient, Visualization-enhanced, and Parallel-accelerated Tool For GWAS    # 
# Version: 1.0.1                                                                           # 
# Authors: Lilin Yin, Haohao Zhang, Zhiwu Zhang, Xinyun Li, Shuhong Zhao, and Xiaolei Liu  # 
# Contact: [email protected]                                                     # 
#------------------------------------------------------------------------------------------# 
[1] "Input data has 378 individuals, 258873 markers"
[1] "Principal Component Analysis Start..."
 means for first 10 snps:
 [1] 0 0 0 0 0 0 0 0 0 0
[1] "GWAS Start..."
[1] "FarmCPU Start ..."
[1] "Current loop: 1 out of maximum of 10"
[1] "seqQTN"
NULL
[1] "scanning..."
[1] "number of covariates in current loop is:"
[1] 8
Error in solve.default(ww) : 
  Lapack routine dgesv: system is exactly singular: U[5,5] = 0

I got an error about Error in solve.default(ww).The kinship and PCA were calculated by MVP.
Please help,Thanks.

Hi Xiaolei,

I am getting the following error

[1] "Current loop: 2 out of maximum of 10"
[1] "Optimizing Pseudo QTNs..."
Error: $ operator is invalid for atomic vectors
In addition: Warning messages:
1: In ID.GP/bin.size :
 longer object length is not a multiple of shorter object length
2: In if (!is.na(inclosure.size)) { :
  the condition has length > 1 and only the first element will be used
3: In parallel::mclapply(1:m, seqQTN.optimize.parallel, mc.cores = ncpus) :
  all scheduled cores encountered errors in user code
Execution halted

I have tried changing minor things, but cannot figure out what the issue is. Any help would be appreciated.

Hi,

I'm running an analysis with rMVP with a HapMap formatted SNP file, and it crashes in the PCA step.
The file was converted to HapMap from VCF via Tassel's export function. The same file works fine in GAPIT.

My analysis script:

library(rMVP)
MVP.Data(fileHMP="input.hmp.txt",
         filePhe="Phenotype.csv",
         fileKin=FALSE,
         filePC=FALSE,
         out="mvp.vcf",
         priority="memory",
)


genotype <- attach.big.matrix("mvp.vcf.geno.desc")

phenotype <- read.table("mvp.vcf.phe",head=TRUE)

map <- read.table("mvp.vcf.map" , head = TRUE)


MVP
for(i in 2:ncol(phenotype)){
  imMVP <- MVP(
    phe=phenotype[, c(1, i)],
    geno=genotype,
    map=map,
    nPC.GLM=5,
    nPC.MLM=3,
    nPC.FarmCPU=3,
    perc=1,
    file='tiff',
    priority="memory",,
    ncpus=12,
    vc.method="EMMA",
    maxLoop=10,
    method.bin="FaST-LMM",#"FaST-LMM","EMMA", "static"
    #permutation.threshold=TRUE,
    #permutation.rep=100,
    threshold=0.05,
    method=c("FarmCPU", 'MLM')
  )
}

The STDOUT:

Preparing data for MVP...
Reading file...
inds: 83        markers:4640569
Preparation for GENOTYPE data is done within 1m54s
83 common individuals between phenotype and genotype.
Preparation for PHENOTYPE data is Done within 0s
Imputing...
Impute Genotype File is done!
MVP data prepration accomplished successfully!

The STDERR:

Loading required package: Rcpp
Loading required package: RcppProgress
Loading required package: BH
Loading required package: bigmemory
Loading required package: biganalytics
Loading required package: foreach
Loading required package: biglm
Loading required package: DBI
Loading required package: parallel
Loading required package: MASS
0%   10   20   30   40   50   60   70   80   90   100%
[----|----|----|----|----|----|----|----|----|----|
**************************************************|
Error in svd(x, nu = 0, nv = k) : infinite or missing values in 'x'
Calls: MVP -> MVP.PCA -> prcomp -> prcomp.default -> svd
Execution halted

Have you seen this before?

Hello Xiaolei

I have a problem while creating kinship files. I have loaded "mvp.geno.desc" as genotype in my R, but when I tries "MVP.Data.Kin(TRUE, mvp_prefix='mvp', out='mvp'", it does not work.

>  MVP.Data.PC(TRUE, out='mvp', perc=1, pcs.keep=5)
Error in MVP.Data.PC(TRUE, out = "mvp", perc = 1, pcs.keep = 5) : 
   could not find function "MVP.Data.PC"

The same thing happened when I typed "MVP.Data.PC(TRUE, out='mvp', perc=1, pcs.keep=5)". How can I fix this problem? Otherwise, may I have your example the kinship file and Principal Components files?

Best regards
Leah

Getting this error after "Visualization Start.... "

Error in seq.default(0, maxbin.num, len) : invalid '(to - from)/by'

错误代码：

devtools::install_github("xiaoleiLiubio/MVP")
Installation failed: Failed to connect to raw.githubusercontent.com port 443: Timed out
devtools::install_github("xiaoleiLiubio/MVP")
Downloading GitHub repo xiaoleiLiubio/MVP@master
from URL https://api.github.com/repos/xiaoleiLiubio/MVP/zipball/master
Installation failed: transfer closed with outstanding read data remaining

Hi,

I tried to load my vcf file using the MVP.Data function to convert it to numeric but I always get the the error

"Error in VCF.Numeralization(x = tt2[-c(1:9)][index], impute = SNP.impute) : object 'maxA' not found"

I hope you can help me.

Cheers,

Stephan

I'm having trouble with my data in MVP that leads to outputs lacking all genotype information per individuals in the GWAS population.

My SNP data was in .tped/.tfam, so I converted it to .vcf with PLINK before loading this data into MVP.Data. Here's the first 10 items from the last line of my .vcf input:
scaffold_724 20159 724_20159 2 1 . . PR GT 0/1

I used this command:

MVP.Data(fileVCF="my_prefix.vcf",
         # used grep in bash and found 1529 header lines with ##, confirmed with head (also bash)
         vcf.jump=1529,
         sep.vcf="\t",
         filePhe="my_pheno_data.txt",
         fileKin=TRUE,
         filePC=TRUE,
         out="my_out")

Everything appeared to be working fine, until I realized when the processing was nearly finished that the output files were only about 200MB combined (and my .vcf input is 28GB). Then, this happened:

... [1] "Number of Markers Written into File: 1: 8252000" [1] "Number of Markers Written into File: 1: 8252993" [1] "Preparation for numeric data is done!" [1] "Preparation for PHENOTYPE data is done!" [1] "Output mvp genotype..." [1] "Calculate KINSHIP using Vanraden method..." [1] "Preparation for Kinship matrix is done!" [1] "Principal Component Analysis Start ..." means for first 10 snps: [1] 0 0 0 0 0 0 0 0 0 0 [1] "Preparation for PC matrix is done!" [1] "MVP data prepration accomplished successfully!" Warning message: In big.PCA(bigMat = big.geno, pcs.to.keep = pcs.keep) : selected too many PCs to keep [5], changing to 2

> Covariates <- attach.big.matrix("my_prefix.pc.desc")
> head(Covariates)
     [,1] [,2]
[1,]    1    0
[2,]    0    1
> head(Kinship)
     [,1] [,2]
[1,]    0    0
[2,]    0    0
--

The genotype file (.map) contains only 3 columns and doesn't show the genotypes for each individual, just the Chr, SNP position and SNP ID.

Thanks for making this tool and for the help!

Hi,

rMVP is a very excellent package and I was wondering whether I can wrap your R package into a Julia package. Then Julia users are able to use your package.

Thanks,