add reps to x_DH

cjGO · Jun 12, 2024 · c0af82e · c0af82e
1 parent 80c50d0
commit c0af82e
Show file tree

Hide file tree

Showing 11 changed files with 67 additions and 186 deletions.
diff --git a/README.md b/README.md
@@ -34,55 +34,38 @@ First, define the genome of your crop
 
 ``` python
 import torch
+```
 
-ploidy = 2
-n_chr = 10
-n_loci = 100
-n_Ind = 3330
-g = Genome(ploidy, n_chr, n_loci)
-population = Population()
-population.create_random_founder_population(g, n_founders=n_Ind)
-init_pop = population.get_dosages().float()  # gets allele dosage for calculating trait values
-
+``` python
+g = Genome()
+founder_pop = Population()
+founder_pop.create_random_founder_population(g, 30)
 # multi_traits
-
-
-target_means = torch.tensor([0, 5])
-target_vars = torch.tensor([1, 1])  # Note: I'm assuming you want a variance of 1 for the second trait
+target_means = torch.tensor([0, 5, 20])
+target_vars = torch.tensor([1, 1, 0.5])  # Note: I'm assuming you want a variance of 1 for the second trait
 correlation_values = [
-        [1.0, 0.8],
-        [0.8, 1.0],
+        [1.0, 0.2, 0.58],
+        [0.2, 1.0, -0.37],
+        [0.58, -0.37, 1.0],
     ]
-
-
-correlated_traits = corr_traits(g, init_pop, target_means, target_vars, correlation_values)
+traits = corr_traits(g, founder_pop.get_dosages().float(), target_means, target_vars, correlation_values)
 ```
 
     Created genetic map
 
 ``` python
-current_pop = init_pop
+f1 = x_random(g, founder_pop.get_genotypes().float(), 50) 
+f1.shape
 ```
 
-``` python
-new_parents = population.get_genotypes().float()
-means = []
-for i in range(10):
-    curr_pop = random_crosses(g, new_parents, 200)
-    new_pheno = correlated_traits[0](curr_pop.sum(dim=1))
-    means.append(new_pheno.mean())
-    topk = torch.topk(new_pheno,10) # select  top 50 parents
-    new_parents = curr_pop.float()[topk.indices]
-```
+    torch.Size([50, 2, 10, 5])
 
 ``` python
-plt.scatter(range(len(means)),means)
+DH = x_DH(g,f1)
 ```
 
-![](index_files/figure-commonmark/cell-6-output-1.png)
-
 ``` python
-plt.scatter(correlated_traits[0](init_pop,h2=.5),correlated_traits[1](init_pop,h2=1))
+DH.shape
 ```
 
-![](index_files/figure-commonmark/cell-8-output-1.png)
+    torch.Size([50, 2, 10, 5])
diff --git a/chewc/__init__.py b/chewc/__init__.py
@@ -1 +1 @@
-__version__ = "0.0.5"
+__version__ = "0.0.6"
diff --git a/chewc/_modidx.py b/chewc/_modidx.py
@@ -33,8 +33,8 @@
                             'chewc.core.PopulationDataset.__init__': ('core.html#populationdataset.__init__', 'chewc/core.py'),
                             'chewc.core.PopulationDataset.__len__': ('core.html#populationdataset.__len__', 'chewc/core.py'),
                             'chewc.core.create_population_dataloader': ('core.html#create_population_dataloader', 'chewc/core.py')},
-            'chewc.cross': { 'chewc.cross.double_haploid': ('cross.html#double_haploid', 'chewc/cross.py'),
-                             'chewc.cross.random_crosses': ('cross.html#random_crosses', 'chewc/cross.py')},
+            'chewc.cross': { 'chewc.cross.x_DH': ('cross.html#x_dh', 'chewc/cross.py'),
+                             'chewc.cross.x_random': ('cross.html#x_random', 'chewc/cross.py')},
             'chewc.crossing': { 'chewc.crossing.double_haploid': ('crossing.html#double_haploid', 'chewc/crossing.py'),
                                 'chewc.crossing.random_crosses': ('crossing.html#random_crosses', 'chewc/crossing.py')},
             'chewc.meiosis': { 'chewc.meiosis.poisson_crossing_over': ('meiosis.html#poisson_crossing_over', 'chewc/meiosis.py'),

diff --git a/chewc/cross.py b/chewc/cross.py
@@ -1,7 +1,7 @@
 # AUTOGENERATED! DO NOT EDIT! File to edit: ../nbs/04_cross.ipynb.
 
 # %% auto 0
-__all__ = ['random_crosses', 'double_haploid']
+__all__ = ['x_random', 'x_DH']
 
 # %% ../nbs/04_cross.ipynb 3
 from .core import *
@@ -11,7 +11,7 @@
 import torch
 
 # %% ../nbs/04_cross.ipynb 4
-def random_crosses( genome: Genome, parent_haplotypes: torch.Tensor, n_crosses: int) -> torch.Tensor:
+def x_random( genome: Genome, parent_haplotypes: torch.Tensor, n_crosses: int) -> torch.Tensor:
     """
     Generate random crosses from a set of parent haplotypes.
 
@@ -49,8 +49,8 @@ def random_crosses( genome: Genome, parent_haplotypes: torch.Tensor, n_crosses:
 
     return progeny_haplotypes
 
-# %% ../nbs/04_cross.ipynb 8
-def double_haploid(genome: Genome, parent_haplotypes: torch.Tensor) -> torch.Tensor:
+# %% ../nbs/04_cross.ipynb 7
+def x_DH(genome: Genome, parent_haplotypes: torch.Tensor) -> torch.Tensor:
     """
     Generate doubled haploid individuals from a set of parent haplotypes.
 

diff --git a/nbs/01_core.ipynb b/nbs/01_core.ipynb
@@ -361,16 +361,16 @@
     {
      "data": {
       "text/plain": [
-       "tensor([[0.0000e+00, 4.4549e+00, 9.7864e+00, 3.2619e+01, 8.7588e+01],\n",
-       "        [0.0000e+00, 1.4553e+01, 4.2347e+01, 6.4445e+01, 8.7742e+01],\n",
-       "        [0.0000e+00, 4.3246e+01, 6.5383e+01, 7.5921e+01, 8.8695e+01],\n",
-       "        [0.0000e+00, 5.7855e+00, 2.0517e+01, 4.6006e+01, 6.2638e+01],\n",
-       "        [0.0000e+00, 3.1555e-02, 2.5189e+01, 7.0929e+01, 9.8765e+01],\n",
-       "        [0.0000e+00, 8.9019e+00, 2.7753e+01, 4.0029e+01, 9.0203e+01],\n",
-       "        [0.0000e+00, 4.3217e+00, 2.2171e+01, 3.2822e+01, 4.1771e+01],\n",
-       "        [0.0000e+00, 1.4086e+01, 4.6586e+01, 8.6157e+01, 9.6368e+01],\n",
-       "        [0.0000e+00, 4.3982e+01, 5.5968e+01, 5.8194e+01, 6.7353e+01],\n",
-       "        [0.0000e+00, 1.4841e+01, 2.9923e+01, 6.3333e+01, 7.7275e+01]])"
+       "tensor([[ 0.0000,  0.2113,  5.1624, 26.2153, 29.5955],\n",
+       "        [ 0.0000,  0.5898, 34.2930, 35.4808, 50.0303],\n",
+       "        [ 0.0000, 35.2943, 39.6551, 44.4670, 99.3925],\n",
+       "        [ 0.0000, 12.7377, 64.0160, 84.8742, 90.3918],\n",
+       "        [ 0.0000, 21.9701, 29.2368, 43.9047, 52.4668],\n",
+       "        [ 0.0000, 31.6351, 31.7786, 40.3927, 42.7500],\n",
+       "        [ 0.0000,  3.6074, 39.6339, 55.3322, 73.2535],\n",
+       "        [ 0.0000, 23.3346, 40.6038, 79.3619, 90.6116],\n",
+       "        [ 0.0000,  7.7061, 46.0459, 69.5689, 72.6099],\n",
+       "        [ 0.0000, 66.2367, 72.1667, 90.1024, 94.9115]])"
       ]
      },
      "execution_count": null,

diff --git a/nbs/02_trait.ipynb b/nbs/02_trait.ipynb
@@ -200,8 +200,6 @@
     "init_pop = population.get_dosages().float()  # gets allele dosage for calculating trait values\n",
     "\n",
     "# multi_traits\n",
-    "\n",
-    "\n",
     "target_means = torch.tensor([0, 5, 20])\n",
     "target_vars = torch.tensor([1, 1, 0.5])  # Note: I'm assuming you want a variance of 1 for the second trait\n",
     "correlation_values = [\n",

diff --git a/nbs/03_meiosis.ipynb b/nbs/03_meiosis.ipynb
@@ -144,56 +144,6 @@
     "    #    import pdb; pdb.set_trace()  # Set the breakpoint"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "bc5fbd18",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Created genetic map\n"
-     ]
-    },
-    {
-     "ename": "NameError",
-     "evalue": "name 'random_crosses' is not defined",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
-      "Cell \u001b[0;32mIn[6], line 23\u001b[0m\n\u001b[1;32m     16\u001b[0m correlation_values \u001b[38;5;241m=\u001b[39m [\n\u001b[1;32m     17\u001b[0m         [\u001b[38;5;241m1.0\u001b[39m, \u001b[38;5;241m0.8\u001b[39m],\n\u001b[1;32m     18\u001b[0m         [\u001b[38;5;241m0.8\u001b[39m, \u001b[38;5;241m1.0\u001b[39m],\n\u001b[1;32m     19\u001b[0m     ]\n\u001b[1;32m     22\u001b[0m correlated_traits \u001b[38;5;241m=\u001b[39m corr_traits(g, init_pop, target_means, target_vars, correlation_values)\n\u001b[0;32m---> 23\u001b[0m \u001b[43mrandom_crosses\u001b[49m(g, init_pop, \u001b[38;5;241m50\u001b[39m)\n",
-      "\u001b[0;31mNameError\u001b[0m: name 'random_crosses' is not defined"
-     ]
-    }
-   ],
-   "source": [
-    "import torch\n",
-    "ploidy = 2\n",
-    "n_chr = 10\n",
-    "n_loci = 100\n",
-    "n_Ind = 3330\n",
-    "g = Genome(ploidy, n_chr, n_loci)\n",
-    "population = Population()\n",
-    "population.create_random_founder_population(g, n_founders=n_Ind)\n",
-    "init_pop = population.get_dosages().float()  # gets allele dosage for calculating trait values\n",
-    "\n",
-    "# multi_traits\n",
-    "\n",
-    "\n",
-    "target_means = torch.tensor([0, 5])\n",
-    "target_vars = torch.tensor([1, 1])  # Note: I'm assuming you want a variance of 1 for the second trait\n",
-    "correlation_values = [\n",
-    "        [1.0, 0.8],\n",
-    "        [0.8, 1.0],\n",
-    "    ]\n",
-    "\n",
-    "\n",
-    "correlated_traits = corr_traits(g, init_pop, target_means, target_vars, correlation_values)\n"
-   ]
-  },
   {
    "cell_type": "code",
    "execution_count": null,

diff --git a/nbs/04_cross.ipynb b/nbs/04_cross.ipynb
@@ -54,7 +54,7 @@
    "outputs": [],
    "source": [
     "#| export\n",
-    "def random_crosses( genome: Genome, parent_haplotypes: torch.Tensor, n_crosses: int) -> torch.Tensor:\n",
+    "def x_random( genome: Genome, parent_haplotypes: torch.Tensor, n_crosses: int) -> torch.Tensor:\n",
     "    \"\"\"\n",
     "    Generate random crosses from a set of parent haplotypes.\n",
     "\n",
@@ -93,14 +93,6 @@
     "    return progeny_haplotypes"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "49898066",
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -126,7 +118,7 @@
     "init_pop = population.get_genotypes().float()  # gets allele dosage for calculating trait values\n",
     "\n",
     "# Convert to numpy array\n",
-    "boolean_array = random_crosses(g, init_pop, 10)"
+    "boolean_array = x_random(g, init_pop, 10)"
    ]
   },
   {
@@ -148,7 +140,7 @@
    "outputs": [],
    "source": [
     "#|export\n",
-    "def double_haploid(genome: Genome, parent_haplotypes: torch.Tensor) -> torch.Tensor:\n",
+    "def x_DH(genome: Genome, parent_haplotypes: torch.Tensor) -> torch.Tensor:\n",
     "    \"\"\"\n",
     "    Generate doubled haploid individuals from a set of parent haplotypes.\n",
     "\n",

diff --git a/nbs/exp1.ipynb b/nbs/exp1.ipynb
@@ -30,7 +30,9 @@
    "id": "98eed592",
    "metadata": {},
    "source": [
-    "use an agent to decide which QTL to gene edit in a given line... plus budget etc... idea is that it will pick clever QTL based not on estimated additive value but also factor in haplotype diversity"
+    "use an agent to decide which QTL to gene edit in a given line... plus budget etc... idea is that it will pick clever QTL based not on estimated additive value but also factor in haplotype diversity\n",
+    "\n",
+    "may be interesting to see if it waits later in the breeding program to select/use budget for gene edits e.g. it waits til later stages of breeding program once the haplotype population structure plays out after generations of mendelian sampling."
    ]
   },
   {