Repository navigation
Expand file tree
/
Copy pathChanges
More file actions
38 lines (36 loc) · 2.31 KB
/
Copy pathChanges
File metadata and controls
38 lines (36 loc) · 2.31 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
Summary of important user-visible changes for Bio-PopGen
--------------------------------------------------------
{{$NEXT}}
* Bio::PopGen::Population->haploid_population now returns new haploid
individuals (one per chromosome, named <id>.1, <id>.2, ...). It used to
return the original diploid individuals twice. Individuals with no
genotype at a marker no longer make it die.
* Bio::PopGen::Statistics: for Bio::PopGen::PopulationI input, the Fu and
Li D, D*, F, F* and Tajima D wrappers now use the number of chromosomes
as the sample size, as theta and segregating_sites_count already did.
Values for diploid populations change; haploid data is unaffected.
* Bio::PopGen::Statistics: mcdonald_kreitman accepts PopulationI objects
(it died with "Not an ARRAY reference"); composite_LD accepts an array
reference of individuals (it dropped them); the $codon_table package
variable is assigned instead of being left undefined.
* Bio::PopGen::Population->add_Individual no longer stores undef or
non-IndividualI values it warned about.
* Bio::PopGen::Simulation::Coalescent: -maxcount is honored by next_tree.
* Bio::PopGen::Simulation::GeneticDrift: next_generation('population')
returns a Bio::PopGen::Population (the class name was misspelled).
* Bio::PopGen::IO->_guess_format recognizes .dat (not only .datX),
.prettybase, .phase and .hap/.hapmap.
* Bio::PopGen::Genotype->get_Alleles compiled the $BlankAlleles pattern
with /o, so the first value seen was used for the whole process. The
hapmap and phase parsers no longer clear $BlankAlleles globally, so N
and ? are treated as missing data in those formats as in the others.
* Bio::PopGen::Statistics: the Fu and Li *_counts forms warn and return
undef when the variance term is zero (n=2, or n=3 for D*) instead of
dying with a division by zero; tajima_D_counts does the same for n=1.
composite_LD and mcdonald_kreitman skip an individual that has no
genotype at a marker instead of dying.
* Bio::PopGen::Individual->get_Genotypes(-marker => $m) returns an empty
list, not a list holding undef, when the individual has no genotype at
that marker. pi and derived_mutations no longer die on such input.
1.7.3 2018-09-11 13:37:12-05:00 America/Chicago
* First release after split from bioperl-live.