Merge pull request #3 from i-pletenev/add_feature_analysis

Add feature analysis

code/07_Fig2_tad_borders_properties_part1.ipynb

@@ -49,8 +49,8 @@
     "boundaries_plus[\"border_num1\"]= range(1,boundaries_plus.shape[0] + 1)\n",
     "boundaries_minus[\"border_num2\"]= range(1,boundaries_minus.shape[0] + 1)\n",
     "\n",
-    "boundaries_minus_sploped = pbt.from_dataframe(boundaries_minus).slop(b=22_500, g=\"../data/chrom.sizes\")\n",
-    "boundaries_plus_sploped = pbt.from_dataframe(boundaries_plus).slop(b=22_500, g=\"../data/chrom.sizes\")"
+    "boundaries_minus_sploped = pbt.from_dataframe(boundaries_minus).slop(b=22_500, g=\"../data/hg38.chrom.sizes\")\n",
+    "boundaries_plus_sploped = pbt.from_dataframe(boundaries_plus).slop(b=22_500, g=\"../data/hg38.chrom.sizes\")"
    ]
   },
   {

code/08_Fig2_tad_and_thier_borders_properties_part2.ipynb

@@ -762,8 +762,8 @@
     }
    ],
    "source": [
-    "boundaries_minus_sploped = pbt.from_dataframe(boundaries_minus.iloc[:, [0,1,2, 5]]).slop(b=15000, g=\"../data/chrom.sizes\")\n",
-    "boundaries_plus_sploped = pbt.from_dataframe(boundaries_plus.iloc[:, [0,1,2, 5]]).slop(b=15000,g=\"../data/chrom.sizes\")"
+    "boundaries_minus_sploped = pbt.from_dataframe(boundaries_minus.iloc[:, [0,1,2, 5]]).slop(b=15000, g=\"../data/hg38.chrom.sizes\")\n",
+    "boundaries_plus_sploped = pbt.from_dataframe(boundaries_plus.iloc[:, [0,1,2, 5]]).slop(b=15000,g=\"../data/hg38.chrom.sizes\")"
    ]
   },
   {
@@ -774,7 +774,7 @@
    "outputs": [],
    "source": [
     "def make_tss(pc_DE_genes_minus):\n",
-    "    chrom_sizes_path = \"../data/chrom.sizes\"\n",
+    "    chrom_sizes_path = \"../data/hg38.chrom.sizes\"\n",
     "    tss = pbt.from_dataframe(pc_DE_genes_minus).flank(g=chrom_sizes_path, s=True, l=1, r=0)\n",
     "    tss = pd.read_table(tss.fn, header = None, names=pc_DE_genes_minus.columns.tolist())\n",
     "    return tss\n",
@@ -904,7 +904,7 @@
    "outputs": [],
    "source": [
     "def make_a_b(expressed_not_de, border_file_init):  \n",
-    "    border_file_slopped = pbt.from_dataframe(border_file_init.iloc[:, [0,1,2, 5]]).slop(b=15000, g=\"../data/chrom.sizes\")\n",
+    "    border_file_slopped = pbt.from_dataframe(border_file_init.iloc[:, [0,1,2, 5]]).slop(b=15000, g=\"../data/hg38.chrom.sizes\")\n",
     "\n",
     "    b_noDE = border_file_slopped\\\n",
     "                  .intersect(pbt.from_dataframe(expressed_not_de), wa=True, wb=True)\n",

code/09_Fig3_chromtain_loops_properties.ipynb

@@ -726,7 +726,7 @@
    "source": [
     "columns = [\"chrom\",\"start\",\"end\",\"gene.name\", \"gene.id\",\"strand\"]\n",
     "columns2 = ['chrom', 'start', 'end', 'gene.name', 'log2FC', 'strand']\n",
-    "chrom_sizes_path = '../data/chrom.sizes' #set path to chrom.sizes\n",
+    "chrom_sizes_path = '../data/hg38.chrom.sizes' \n",
     "tss_plus = pbt.from_dataframe(de_neurons).flank(g=chrom_sizes_path, s=True, l=2000, r=0)\n",
     "tss_plus = pd.read_table(tss_plus.fn, header = None, names=columns2)\n",
     "\n",
@@ -756,7 +756,7 @@
     "    df = pd.concat([df_sel[[\"chrom1\",\"start1\", \"end1\"]],\n",
     "          df_sel[[\"chrom2\",\"start2\", \"end2\"]].rename(columns = {\"chrom2\":\"chrom1\",\"start2\":\"start1\",\"end2\":\"end1\"})])\n",
     "    df = df.drop_duplicates().reset_index(drop=True)  \n",
-    "    df =pbt.from_dataframe(df).slop(b=slop, g=\"../data/chrom.sizes\")\n",
+    "    df =pbt.from_dataframe(df).slop(b=slop, g=\"../data/hg38.chrom.sizes\")\n",
     "    \n",
     "    colnames = de_neurons.columns.tolist()    \n",
     "    df_intersect = pbt.from_dataframe(de_neurons)\\\n",
@@ -890,7 +890,7 @@
     "    df_processed = df[columns].reset_index(drop=True)\n",
     "    df_processed[\"anchor\"] = anchor    \n",
     "    if slopvalue and slopvalue > 0:\n",
-    "        df_slopped = pbt.from_dataframe(df_processed).slop(b=slopvalue, g='../data/chrom.sizes')\n",
+    "        df_slopped = pbt.from_dataframe(df_processed).slop(b=slopvalue, g='../data/hg38.chrom.sizes')\n",
     "        df_processed = pd.read_table(df_slopped.fn, header=None, names=df_processed.columns.tolist())\n",
     "        assert df_processed.shape[0] > 1, \"Dataframe after slopping has less than 2 rows.\"\n",
     "    \n",