Tables and Figures from the Anthropology and Human diversity component of the 13th IHW
For additional information see: Single et al. (2004). In "Haplotype Frequencies and Linkage Disequilibrium among classical HLA genes". Hansen, J. and Dupont, B. (Eds). IHWG Press, Seattle.
Figures
-
Results for each of the following locus pairs are reported in a separate graph:
A:B, C:B, B:DRB1, A:DPB1, B:DPB1, DRB1:DQB1, DRB1:DBP1. In each graph the
extent of haplotype sharing among regions is summarized by two columns. The
left column for each region depicts the fraction of haplotypes in a region
which are either exclusively found in that region (black), shared by two (dark
gray), three (light gray), or more than three regions (white) regions. The
second column shows the contribution of the haplotypes in each category to the
overall haplotype frequency in that region (i.e., a frequency weighted version
of the first column). The total number of haplotypes is indicated at the top of
each set of columns.
HF-regional-sharing.pdf (pdf document)
Tables
- Summary tables (Table 2 in Single et al. (2004)). These contain statistics
that summarize the distribution of haplotypes by region and across all regions.
Each haplotype is presented in a separate worksheet, and the legends are provided
in the first worksheet.
HF-summary.xls (excel document)
Go to detailed description of Summary tables
- Regional tables (Table 3 in Single et al. (2004)). These allow an
assessment of the degree of sharing of specific alleles found at or above the
threshold frequency of 0.05 in at least one population sample (regionally
frequent haplotypes). The statistic reported in the tables is the maximum
frequency for that haplotype in samples from the region.
Each locus pair is presented in a separate worksheet. The legend is provided
in the first worksheet.
HF-regional.xls (excel document)
Go to detailed description of Regional tables
- Population tables These tables provide a complete listing of estimated
frequencies for all haplotypes that are inferred to be present in the sample as
at least one copy. Each locus is in a separate file, and within each file
each region is is presented in a separate worksheet. These files extend those reported in
Single et al. (2004), in which only a subset of alleles were reported (those that were
among the three most frequent in any population). The legend is is provided in
the first worksheet.
HF-population_DRB1-DQB1.xls
HF-population_DRB1-DQA1-DQB1.xls
HF-population_DRB1-DPB1.xls
HF-population_C-B.xls
HF-population_B-DRB1.xls
HF-population_B-DPB1.xls
HF-population_A-DPB1.xls
HF-population_A-C-B.xls
HF-population_A-B.xls
HF-population_A-B-DRB1.xls
Detailed description of table contents
- Summary table
The summary table contains statistics that summarize the distribution of
haplotypes by region as well as across all regions. The first column
corresponds to all regions combined and is followed by a separate column for
each region. In this table we report sample-size weighted averages over all
populations in each region where it is appropriate. Due to the presence of a
few very large samples, the following procedure was used to modify the weights
used. If the maximum sample size in a region was more than 1.5 times the next
largest in that region, the weight used for the largest population was 1.5
times that of the next largest. Below is a description of each statistic
presented.
- Number of populations typed
- Expected heterozygosity (gene diversity) at the haplotype level.
- Number of different haplotypes (k) per region: This is the number of
haplotypes estimated with at least one copy in the dataset.
- Range of values for k per population: The minimum and maximum values of k
for populations in the region are reported.
- Number of different haplotypes needed to reach a frequency of 0.5: This is
the minimum number of different haplotypes necessary to account for 50% of the
haplotype frequency. Small values correspond to datasets where a few haplotypes
are present at high frequencies; high values are observed in cases where
haplotype frequencies are more even, with several haplotypes at similar
frequencies at the high end of the spectrum.
- Number of regionally frequent haplotypes: This is the number of
haplotypes with a frequency of 0.05 or more which are present in at least one
population sample.
- Regionally frequent haplotypes that are frequent in only one
region: For each region, this specifies how many of the frequent
haplotypes have a frequency at or above the threshold exclusively in that
region.
- Fraction of haplotypes estimated with counts equal to one: This is the
number of singletons divided by the number of different haplotypes. It
indicates how many of the observed haplotypes are at the lowest end of the
frequency spectrum.
- Total frequency of haplotypes estimated with count equal to one: This
statistic provides additional information about the low end of the frequency
spectrum by giving the total frequency contribution of singletons.
- Wn and D': These are measures of overall LD
- Complementary Fraction: This is the fraction of haplotypes for which
Dij' = 1.
- FNF: This is the fraction of the number of haplotypes expected under
linkage equilibrium that are not found in the sample.
- Regional Table
This table allows an assessment of the degree of sharing of specific haplotypes
found at or above the threshold frequency of 0.05 in at least one population
sample regionally frequent haplotypes). The statistic reported in the
tables is the maximum frequency for that haplotype in samples from the region.
While other statistics could have been used for these tables (mean, median,
etc.), we chose the maximum to highlight the sharing (and absence from some
regions) of specific high frequency haplotypes.
Haplotypes that were observed in only one population sample in that region are
marked with an asterisk, in order to highlight regionally frequent
haplotypes that were not shared within the region. This table facilitates the
identification of specific haplotypes responsible for the patterns seen in the
summary table. The number of haplotypes listed is equal to the total number of
regionally frequent haplotypes indicated in the summary table.