update

Author: stsmall

aintrogression.py

@@ -49,9 +49,9 @@
 
 R_D: (Racimo 2016)
 
-U20: (Racimo 2016, Jagoda 2018)
+Utwenty: (Racimo 2016, Jagoda 2018)
 
-Q95: (Racimo 2016, Jagoda 2018)
+QninetyFive: (Racimo 2016, Jagoda 2018)
 
 Dp_intro: (Racimo 2016)
 
@@ -175,12 +175,64 @@ def QninetyFive(pos, gt, out, target, bait, clen, ofreq=0.01, winSize=10000, win
     return(Q95)
 
 
-def Fd(pos, gt, out, target, bait, ofreq=0.01, winSize=10000, winStep=None):
+def F_d(pos, gt, out, target, bait, ofreq=0.01, winSize=10000, winStep=None):
     """Fd from Martin 2015
     """
     return(None)
 
 
+def Dout():
+    """
+    dout: (dXO + dYO) / 2
+    average distance between species X and the Outgroup and Species Y and the
+    Outgroup.
+    """
+
+
+def Dxy():
+    """
+    dxy: pairwise distance / bases; low values possible introgression
+    dxy as number of sequence diff between any 2 sequences, x and y, in two
+    taxa, X and Y (divided by the number of sites), then dxy is the average
+    distance between all sequences in the two species. # above assumes no
+    variation in neutral mutation rate, low mutation rate can be mistaken for
+    recent introgression
+    """
+
+
+def Dmin():
+    """
+    REQUIRES PHASES, HAPLOTYPEARRAY
+    dmin: min(dxy), requires haplotypes
+    minimum distance among all pairing of haplotypes in the 2 species. Pvalue
+    by coalescent with no migration or from other parts of the genome average.
+    # above assumes no variation in neutral mutation rate, low mutation rate
+    can be mistaken for recent introgression
+    """
+
+
+def RND():
+    """
+    RND (relative node depth): dxy / dout
+     Robust to low mutation rates like HKY test if neutrality. # not sensitive
+     to low-frequency migrants.
+    calculate Dxy in windows on haplotypeArray between Species X and Y
+    calculate Dxy between Species X and Outgroup
+    calculate Dxy between Species Y and Outgroup
+    """
+#    Dxy/ Dout
+
+
+def RNDmin():
+    """
+    RNDmin: dmin / dout
+    Similarly, like both dmin and Gmin, RNDmin should be sensitive to even rare
+    migrant haplotypes. In addition, we expect RNDmin to be powerful even when
+    migrants are high in frequency
+    """
+#    Dmin/Dout
+
+
 def adaptPlot(Q, chrom, name, p, save=True):
     """
     """

ald.py

@@ -6,8 +6,10 @@
 @author: scott
 """
 import allel
+import matplotlib as mpl
 import matplotlib.pyplot as plt
 import seaborn as sns
+mpl.rcParams['pdf.fonttype'] = 42
 import numpy as np
 import scipy
 from collections import defaultdict

aplot.py

@@ -7,8 +7,10 @@
 """
 import allel
 import numpy as np
+import matplotlib as mpl
 import matplotlib.pyplot as plt
 import seaborn as sns
+mpl.rcParams['pdf.fonttype'] = 42
 
 
 def plotvars(chrm, callset, window_size=10000, title=None, saved=False):

autil.pyc

@@ -92,17 +92,14 @@ def loadgenome_extradata_fx(fasta_handle, gff3_handle, meta):
     genome, gff3, meta = loadgenome_extradata_fx(fasta_handle, gff3_handle,
                                                  meta)
 
-    #meta = "/home/scott/Documents/Wb/Wb_sWGA/data_files/allel/WbAllpops.47.info"
-    meta = "/home/scott/Desktop/AnopSG_liftvcf/AnopSG.55.info"
+    meta = "AnopSG.55.info"
     meta = pd.read_csv(meta, delimiter=",")
-    #var = Chr('All', 'WbAllpops.impute.allchr.47.h5')
-    var = Chr('All', '/home/scott/Desktop/AnopSG_liftvcf/2L.SNP.recode.h5')
+    var = Chr('All', '2L.FSG.SNP.recode.h5')
     popdict = autil.subpops(var, meta, bypop=True, bykary=False)
     pop2color = autil.popcols(popdict)
     chrlist = np.unique(var.chrm[:])
-    # chrlist = del chrlist['Wb_ChrX_0']
     chrlen = {}
-    with open("/home/scott/Documents/Wb/Wb_sWGA/data_files/allel/chr_info", 'r') as c:
+    with open("chr_info", 'r') as c:
         for line in c:
             x = line.strip().split()
             chrlen[x[0]] = int(x[1])
@@ -202,11 +199,20 @@ def loadgenome_extradata_fx(fasta_handle, gff3_handle, meta):
                                  pop2color, var)
 
 
-
-
+import matplotlib.pyplot as plt
+# tajd histogram
 x = []
 for p in tajddict.keys():
-    x.append((tajddict[p]["Haiti"][2][0]))
+    x.append((tajddict[p]["Fun"][2][0]))
 m = np.concatenate(x).ravel()
 n = m[~np.isnan(m)]
 b,bins,patches = plt.hist(n, 50, density=True)
+
+# pi histogram
+x = []
+for p in pidict.keys():
+    x.append((pidict[p]["Par"][2][0]))
+m = np.concatenate(x).ravel()
+n = m[~np.isnan(m)]
+b, bins, patches = plt.hist(n, 50, density=True)
+

run_main.py

@@ -0,0 +1,90 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""
+Created on Mon Jan 21 16:34:02 2019
+
+@author: scott
+"""
+from __future__ import division
+from __future__ import print_function
+import numpy as np
+import pandas as pd
+import matplotlib as mpl
+# functions
+from allel_class import Chr
+import autil as autil
+import adiv as av
+import ald as ald
+import adxy as adxy
+import aplot as aplot
+mpl.rcParams['pdf.fonttype'] = 42
+
+meta = "AnopSG.55.info"
+meta = pd.read_csv(meta, delimiter=",")
+var = Chr('All', '2L.FSG.SNP.recode.h5')
+popdict = autil.subpops(var, meta, bypop=True, bykary=False)
+pop2color = autil.popcols(popdict)
+
+# chrlist = np.unique(var.chrm[:])
+chrlen = {}
+with open("chr_info", 'r') as c:
+    for line in c:
+        x = line.strip().split()
+        chrlen[x[0]] = int(x[1])
+
+# RND
+dxydict = {}
+for c in chrlen.keys():
+    var = Chr('All', '{}.FSG.SNP.recode.h5'.format(c))
+    var.geno(c, meta)
+    # var.miss(var.gt, var.pos, .20)
+    # var.mac(var.gt, var.pos, 1)
+    print("\nStats for Chromosome {}\n".format(c))
+    # allele count object
+    ac_subpops = var.gt.count_alleles_subpops(popdict, max_allele=2)
+    df_dxy = adxy.pairDxy(c, chrlen[c], ac_subpops, var.pos, plot=True)
+    dxydict[c] = df_dxy
+
+# Diversity statistics
+pidict = {}
+tajddict = {}
+thetadict = {}
+for c in chrlen.keys():
+    var = Chr('All', '{}.FSG.SNP.recode.h5'.format(c))
+    var.geno(c, meta)
+    print("\nStats for Chromosome {}\n".format(c))
+    # var.miss(var.gt, var.pos, .20)
+    # var.mac(var.gt, var.pos, 1)
+    # allele count object
+    ac_subpops = var.gt.count_alleles_subpops(popdict, max_allele=1)
+    pi = av.pi(c, chrlen[c], ac_subpops, var.pos, plot=True)
+    pidict[c] = pi
+    d = av.tajd(c, chrlen[c], ac_subpops, var.pos, plot=True)
+    tajddict[c] = d
+    t = av.theta(c, chrlen[c], ac_subpops, var.pos, plot=True)
+    thetadict[c] = t
+
+# Diversity boxplot; sumdict() autil then boxplot in aplot
+
+theta = autil.catdict(thetadict)
+aplot.divboxplot(theta, pop2color)
+pi = autil.catdict(pidict)
+aplot.divboxplot(pi, pop2color)
+tajd = autil.catdict(tajddict)
+aplot.divboxplot(tajd, pop2color)
+
+# tajd histogram how do I get colors and transparency??
+for pop in popdict.keys():
+    b, bins, patches = mpl.pyplot.hist(tajd[pop], 50, density=True)
+
+# LD decay plot
+lddict = {}
+for c in chrlen.keys():
+    var = Chr('All', '{}.FSG.SNP.recode.h5'.format(c))
+    var.geno(c, meta)
+    print("\nStats for Chromosome {}\n".format(c))
+    var.miss(var.gt, var.pos, .20)
+    # allele count object
+    ac_subpops = var.gt.count_alleles_subpops(popdict, max_allele=1)
+    lddict[c] = ald.ld_decay(c, chrlen[c], ac_subpops, popdict,
+                             pop2color, var)