Skip to main content

Data Exploration

Allele frequency table​

allelefreqtable(data::PopData; by::Union{String, Symbol} = "global")

Return a table of the observed global (default) or population allele frequencies in a PopData object. Use this if you want to see what the frequencies are for every allele at every locus.

julia> cats = @nancycats ;

julia> allelefreqtable(cats)
108×4 DataFrame
 Row │ locus   allele  count  frequency  
     │ String  Int16?  Int64  Float64    
─────┼───────────────────────────────────
   1 │ fca8       135    105  0.241935
   2 │ fca8       143     44  0.101382
   3 │ fca8       133     33  0.0760369
   4 │ fca8       137     83  0.191244
  ⋮  │   ⋮       ⋮       ⋮        ⋮
 105 │ fca37      226      2  0.00421941
 106 │ fca37      216      7  0.0147679
 107 │ fca37      224      2  0.00421941
 108 │ fca37      204      6  0.0126582
                         100 rows omitted

Genotype frequency table​

genofreqtable(data::PopData; by::Union{String, Symbol} = "global")

Return a table of the observed global (default) or population genotype frequencies in a PopData object. Use this if you want to see what the frequencies are for every genotype at every locus.

julia> cats = @nancycats ;

julia> genofreqtable(cats)
341×4 DataFrame
 Row │ locus   genotype    count  frequency  
     │ String  Tuple…      Int64  Float64    
─────┼───────────────────────────────────────
   1 │ fca8    (135, 143)     16  0.0737327
   2 │ fca8    (133, 135)      9  0.0414747
   3 │ fca8    (135, 135)     23  0.105991
   4 │ fca8    (137, 143)      8  0.0368664
  ⋮  │   ⋮         ⋮         ⋮        ⋮
 338 │ fca37   (206, 220)      1  0.00421941
 339 │ fca37   (208, 218)      1  0.00421941
 340 │ fca37   (184, 184)      3  0.0126582
 341 │ fca37   (208, 210)      3  0.0126582
                             333 rows omitted

Missing Data​

missingdata(data::PopData; by::Union{String, Symbol} = "sample")

Get missing genotype information in a PopData object. Specify a mode of operation using the by = keyword to return a table corresponding with that missing information type.

bywhat it does
"sample"returns a count of missing loci per individual (default)
"population"returns a count of missing genotypes per population
"locus"returns a count of missing genotypes per locus
"locusxpopulation"returns a count of missing genotypes per locus per population
julia> sharks = @gulfsharks ;

julia> missingdata(sharks)
212×2 DataFrame
 Row │ name     missing
     │ String   Int64
─────┼──────────────────
   1 │ cc_001       124
   2 │ cc_002        94
   3 │ cc_003       100
   4 │ cc_005         0
   5 │ cc_007         2
   6 │ cc_008         1
   7 │ cc_009         2
  ⋮  │    ⋮        ⋮
 206 │ seg_025        0
 207 │ seg_026        0
 208 │ seg_027        2
 209 │ seg_028       25
 210 │ seg_029        0
 211 │ seg_030        1
 212 │ seg_031        1
        198 rows omitted

Pairwise Identical Genotypes​

While not a substitute for a kinship analysis, it may be useful to know or verify how similar your data are in a very literal sense: how many identical genotypes do two individuals have across all loci? To do this, we use pairwiseidentical() to perform an all x all comparison of identical genotypes. This can be done for all individuals in a PopData object, or restricted to a specific set of individuals:

julia> cats = @nancycats;

julia> pairwiseidentical(cats)
27966×4 DataFrame
   Row │ sample_1  sample_2  identical  n     
       │ String    String    Float64    Int64 
───────┼──────────────────────────────────────
     1 │ N215      N216           0.5       8
     2 │ N215      N217           0.25      8
     3 │ N215      N218           0.38      8
     4 │ N215      N219           0.38      8
   ⋮   │    ⋮         ⋮          ⋮        ⋮
 27963 │ N297      N290           0.29      7
 27964 │ N281      N289           0.25      8
 27965 │ N281      N290           0.43      7
 27966 │ N289      N290           0.14      7
                            27958 rows omitted

Allelic Richness​

If you were curious about allelic richness (number of alleles per locus), then you can use richness() to find that out. Use by = "population" to return a table by locus by population.

julia> cats = @nancycats;

julia> richness(cats)
9×2 DataFrame
 Row │ locus   richness 
     │ String  Int64    
─────┼──────────────────
   1 │ fca8          16
   2 │ fca23         11
   3 │ fca43         10
   4 │ fca45          9
   5 │ fca77         12
   6 │ fca78          8
   7 │ fca90         12
   8 │ fca96         12
   9 │ fca37         18

Average Number of Alleles​

Similar to richness, if you wanted to know the average number of alleles per locus, use alleleavg(). Use rounding = false if you don't want the answer rounded to 4 decimal places.

julia> alleleavg(@nancycats)
(mean = 12.0, stdev = 0.2668)

julia> alleleavg(@nancycats, rounding = false)
(mean = 12.0, stdev = 0.2667968432263687)

Summary Statistics​

Population genetics famously includes all manner of coefficients with which to summarize data. Use summary() to view FST, DST, HT, etc. (like Hierfstat::basic.stats). 

julia> summary(@nancycats)
1×10 DataFrame
 Row │ Het_obs  HS       HT       DST      HT′      DST′     FST      FST′     FIS      DEST
     │ Float64  Float64  Float64  Float64  Float64  Float64  Float64  Float64  Float64  Float64
─────┼──────────────────────────────────────────────────────────────────────────────────────────
   1 │  0.6299   0.7083   0.7717   0.0634   0.7757   0.0674   0.0821   0.0869   0.1108    0.231
prime symbol

The column names above use the unicode prime symbol ′ to better reflect the actual coefficient ("FST prime" etc.). To print that character, press \prime<TAB>, which reads "backslash, the word 'prime', and the TAB button".