The files in this directory are used by the UCSC Genome Browser hub
relevant to the paper: "Mapping gene flow between ancient hominins through
demography-aware inference of the ancestral recombination graph" by Hubisz,
Williams, Siepel (biorxiv 2019).

Each directory contains results for a genome-wide run of ARGweaver:
M1A: This is the main set of results presented in the paper, where the
   data set contains 2 African individuals from the SGDP data set
   (Khomani_San_1 and Mandenka_2), two Neanderthals, the Denisovan, and
   a chimpanzee. The migration time (t_mig) was set to 250kya and the
   super-archaic divergence time (t_div) to 1Mya

M2B: This is an alternate set of results with the same data set as M1A,
   but using t_mig=150kya and t_div=1.5Mya. These results are presented
   in the Supplementary Text

ooaM1A: This is a set using a model similar to the one in M1A, but with
   additional migration bands from Nea and Den into humans at 50kya. The
   data included the same individuals as the previous set, but with the
   addition of a Papuan individual (Papuan_1) and a Basque individual
   (Basque_2)

papOOA: This run had the same data as the previous run, except without the
   Basque individual. It was analyzed using a simple model that only
   has migration bands from Neanderthal and the Denisovan into humans 50kya.
   These results were presented in the main text (Figure 4)

fraOOA: This run is similar to papOOA, except it uses the Basque individual
   instead of the Papuan, and were also presented in Figure 4 of the main
   text.


Within each directory, there are several types of files, but the
main ones of interest are the "bw" files. These are bigWig files (UCSC
Genome browser format). They give a probability of introgression per
base for each individual and each type of migration event, as indicated
by the file name. Within the file name, the following codes are used:

het = heterozygous introgressed
hom =  homozygous introgressed
either = heterozygous or homozygous introgressed
alt = Altai Neanderthal
vin = Vindija Neanderthal
den = Denisovan
afr1 = Khomani_San_1
afr2 = Mandenka_2
aToN = Africa to Neanderthal introgression
aToD = Africa to Denisovan introgression
sToA = super-archaic to African
sToN = super-archaic to Neanderthal
sToD = super-archaic to Denisovan
dToH = Denisovan to human
dToN = Neanderthal to human
hToN = Human to Neanderthal
hToD = human to Denisovan

Note that sometimes I use a or A for "African", and others I use h or H for
humans, but the models are the same. It is just depends on whether the data
analyzed included non-African humans.

Other files are likely of less interest; the bed files are summaries of
introgressed regions using a 0.5 cutoff, and the bb (bigBed) files are
conversions of these used by the Genome Browser.

The variantBeds subdirectories contain bigBeds which give the location of
variants for each phased haplotype (using the phasing from the final
ARGweaver-D iteration).