Q-ROADTRIPS (beta) 1.0 is a C program that performs single-SNP, quantitative trait association testing in samples with partially or completely unknown population and/or pedigree structure. Q-ROADTRIPS is an extension of the ROADTRIPS case-control method to quantitative traits: Thornton T., McPeek M. S. "ROADTRIPS: Case-Control Association Testing with Partially or Completely Unknown Population and Pedigree Structure" (2010) American Journal of Human Genetics, vol 86, pp. 172-184. Q-ROADTRIPS uses an empirical covariance matrix calculated from genomewide SNP data to correct for unknown population and pedigree structure.

Q- ROADTRIPS is suitable for applications such as correcting for partially or completely un-known population structure and/or relatedness in the context of GWAS. The Q-ROADTRIPS program can be used for the analysis of any quantitative trait and there are no restrictions on the types of distributions allowed. For example, a trait can follow a normal distribution, can be count values (e.g., from a poisson distribution), a mixture of different distributions, or the distribution of the trait can be completely unknown. Analysis can be performed genomewide (currently just for autosomes).

Please read the documentation file "Q-ROADTRIPS_Documentation.pdf" carefully before using the software. Below is a quick start guide for using the Q-ROADTRIPS software. There are commond line options available that may be useful for the user that are not given below, such as saving the empirical correlation matrix calculatd by the software to a file, with the "-f" option, and an option allowing the user to specify a file containing an empirical correlation matrix for calculating the association test statistic and bypass calculating the empirical correlation matrix from the input genotype data file, with the "-e" option.

The current version of Q-ROADTRIPS is (beta) release 1.0. Q-ROADTRIPS support PLINK's transposed PED file.

Instructions

1. Use PLINK software to transpose your genotype files (bed or ped format) to tped/tfam format by running % plink --bfile [bed_prefix] (or --file [ped_prefix]) --recode12 --output-missing-genotype 0 --transpose --out [tped_prefix]

2. Reformat the phenotype files in the same order of .tfam files. The phenotype file has six entries at each line in the following order: Family ID, Individual ID, Father ID, Mother ID, Sex, and Phenotype. Farther ID, Mother ID, and Sex values will be ignored by the program, however, some string must be present for these columns. Missing phenotype values should be represented as "NA". We recommend using phenotypes that have been adjusted for any relevant covariants, e.g., regress out covariates for the phenotypes.

Sample lines of phenotype files. (tab or space delimited)

59811   8598110	0   0   2   0.0735  
862311  862311	0   0	2   1.309 
864111	864111	0   0	1   -1.54
865211	865211	0   0	1   0.23982
875511	875511	0   0	2   NA
880111	880111	0   0	1   0.1148
880811	880811	0   0	2   0.423
881511	881511	0   0	2   -1.123
88211	88211	0   0	5   2.456

3. To run Q-ROADTRIPS with a PLINK tped file named "myfile.tped" and phenotype file named "myfile.pheno", the following commands can be used.

This will generate the following files:

• QROADTRIPStest.pvalues: lists the p-values for every SNP for the QR test statistic.
• QROADTRIPStest.testvalues: lists, for every SNP, the values of the QR test statistic.
• QROADTRIPStest.top lists the top 20 SNPs with the smallest p-values for QR test.
• QROADTRIPStest.out: this is a detailed output file containing:
  • Summary of the phenotype file information: total number of individuals in the phenotype file, number of individuals with non-missing phenotype values, and number of individuals with no available phenotype information.
  • For each marker
    • SNP identifier/rs number
    • the numbers of genotyped individuals at the marker who have non-missing phenotype values.
    • value of the QR statistic and corresponding p-value using the chi-squared null distribution.
    • a warning message is printed when some allele counts are small, a situation in which the chi-squared asymptotic null distribution might not provide accurate p-values
    • allele frequencies estimated based on allele counts in the sample
• QROADTRIPStest.err: an error file that may contain warnings

5. Please email to tathornt@uw.edu for any further questions. Enjoy!