The IBD analysis option has two main sets of output files: the IBD files and the Kinship files. These files contain the IBD probabilities and the kinship coefficients (both of which are discussed in detail below) for all pairs of indicated individuals and at all requested positions. The individuals to be reported are indicated by marking them as 'affected' using batch items #16 & #17; recall it is easy to indicate that all pairs of all individuals should be reported on by setting batch item #17 to -1. The IBD and kinship values will always be reported at the marker loci. In addition, the kinship coefficients and, when there is no inbreeding, many of the IBD probabilities will be reported at the positions between the markers indicated using batch item #49. These may be either a uniform grid from the first marker up to the last, or a fixed number of points within each marker interval. Finally, users can specify if they want only the IBD output files, only the Kinship output files, or both, by using batch item #39.
The main Kinship output file is titled IKEF-nn.ALL and will often be quite large. This file contains the conditional kinship coefficients (also known as, empirical kinship coefficients) and the simple IBD probabilities for all pedigrees. IKEF stands for IBD and Kinship Exchange Format. This file is designed to make it easy to export all the IBD and kinship data to programs that employ variance components or other methodology to study linkage to quantitative trait loci (QTL), e.g., MENDEL, QTDT and SOLAR. If a requested pair of individuals is not listed in this file, then all conditional kinship coefficients for that pair of individuals are 0 (zero). There is also a file titled IKKEY-nn.TXT that contains a key to some of the labels in the IKEF file.
The IBD-nn.mmm files contain the IBD sharing probabilities estimated within each original pedigree. For each requested pair of individuals, the standard 0, 1, and 2 allele sharing probabilities are reported. These Z0, Z1, and Z2 values, along with the empirical kinship coefficients, are supplied at each marker locus. In addition, the kinship coefficients and, when there is no inbreeding, these allele sharing probabilities are provided at the user-specified points between the marker loci. For each pair of requested individuals, the theoretical kinship coefficient is also reported.
The IBD09-nn.mmm files contain more detailed IBD sharing probabilities estimated within each original pedigree. For each pair of requested individuals, the probabilities of the 9 condensed identity states are reported. Each of the 9 condensed identity states is a different configuration showing which of the four alleles are IBD. Explicitly, if the two affecteds have unordered genotypes (I1,I2) and (J1,J2), then the 9 condensed identity states can be diagrammed as:
S1 => {I1, I2, J1, J2} i.e. all alleles are IBD
S2 => {I1, I2}{J1, J2} i.e. I1 & I2 are IBD and J1 & J2 are IBD
S3 => {I1, I2, J1}{J2} i.e. all alleles are IBD except J2
S4 => {I1, I2}{J1}{J2} .
S5 => {I1}{I2, J1, J2} .
S6 => {I1}{I2}{J1, J2} .
S7 => {I1, J1}{I2, J2}
S8 => {I1}{I2, J1}{J2}
S9 => {I1}{I2}{J1}{J2} i.e. no alleles are IBDThe relationship between the probabilities of the standard 0, 1, and 2 allele sharing states and the condensed identity states is:
Z0 = S2 + S4 + S6 + S9
Z1 = S3 + S5 + S8
Z2 = S1 + S7Note that when a pedigree is not consanguineous S1, S2, S3, S4, S5 and S6 are all zero. Thus, the additional information contained in the condensed identity states compared to the standard 0, 1 and 2 allele sharing is only relevant for consanguineous pedigrees.
The IBD15-nn.mmm files contain the most detailed IBD sharing probabilities estimated within each original pedigree. For each pair of requested individuals, the probabilities of the 15 detailed identity states are reported. Each of the 15 detailed identity states is a different configuration showing which of the four alleles are IBD. Explicitly, if the two affecteds have ordered genotypes (IM,IP) and (JM,JP), then the 15 detailed identity states can be diagrammed as:
S*01 => {IM, IP, JM, JP} i.e. all alleles are IBD
S*02 => {IM, IP, JM}{JP} i.e. all alleles are IBD except JP
S*03 => {IM, IP, JP}{JM} .
S*04 => {IM, JM, JP}{IP} .
S*05 => {IM}{IP, JM, JP} .
S*06 => {IM, IP}{JM, JP}
S*07 => {IM, IP}{JM}{JP}
S*08 => {IM}{IP}{JM, JP}
S*09 => {IM, JM}{IP, JP}
S*10 => {IM, JM}{IP}{JP}
S*11 => {IM}{IP, JP}{JM}
S*12 => {IM, JP}{IP, JM}
S*13 => {IM, JP}{IP}{JM}
S*14 => {IM}{IP, JM}{JP}
S*15 => {IM}{IP}{JM}{JP} i.e. no alleles are IBDThe relationship between the probabilities of the condensed identity states and the detailed identity states is:
S1 = S*01
S2 = S*06
S3 = S*02 + S*03
S4 = S*07
S5 = S*04 + S*05
S6 = S*08
S7 = S*09 + S*12
S8 = S*10 + S*11 + S*13 + S*14
S9 = S*15The relationship between the probabilities of the standard 0, 1, and 2 allele sharing states and the detailed identity states is:
Z0 = S*06 + S*07 + S*08 + S*15
Z1 = S*02 + S*03 + S*04 + S*05 + S*10 + S*11 + S*13 + S*14
Z2 = S*01 + S*09 + S*12Note that when a pedigree is not consanguineous S*01, S*02, S*03, S*04, S*05, S*06, S*07 and S*08 are all zero. The additional information contained in the detailed identity states compared to the standard 0, 1 and 2 allele sharing is particularly relevant for identifying specific inheritance patterns, e.g., imprinting.
Finally, the XOVER-nn.ALL file contains the observed and expected number
of crossovers found in this analysis. Also, the resulting p-value for each marker
interval recombination fraction is reported.