debug profile

bean/cli/profile.py

@@ -32,9 +32,8 @@ def main(args):
             control_condition=args.control_condition,
             max_editing_window_length=args.window_length,
             pam_col=args.pam_col,
+            save_fig=args.save_fig,
         ),
         kernel_name="bean_python3",
     )
-    os.system(
-        f"jupyter nbconvert --to html {args.output_prefix}_editing_preference.ipynb"
-    )
+    os.system(f"jupyter nbconvert --to html {args.output_prefix}.ipynb")

bean/cli/qc.py

@@ -53,4 +53,4 @@ def main(args):
         ),
         kernel_name="bean_python3",
     )
-    os.system(f"jupyter nbconvert --to html {args.out_report_prefix}.ipynb")
+    os.system(f"jupyter nbconvert --to=html {args.out_report_prefix}.ipynb")

bean/notebooks/profile_editing_preference.ipynb

@@ -56,7 +56,8 @@
     "control_condition = \"bulk\"\n",
     "output_prefix = \"editing pattern\"\n",
     "max_editing_window_length = 6\n",
-    "pam_col=\"5-nt PAM\""
+    "pam_col=\"5-nt PAM\"\n",
+    "save_fig=False"
    ]
   },
   {
@@ -617,7 +618,7 @@
     "    nonref_base_changes = None\n",
     ")\n",
     "profile_df.to_csv(f\"{output_prefix}_behive_like_profile.csv\")\n",
-    "plt.savefig(f\"{output_prefix}_behive_like_profile.pdf\", bbox_inches = 'tight')"
+    "if save_fig: plt.savefig(f\"{output_prefix}_behive_like_profile.pdf\", bbox_inches = 'tight')"
    ]
   },
   {
@@ -975,7 +976,7 @@
     "        lw=2\n",
     "    )\n",
     ")\n",
-    "plt.savefig(f\"{output_prefix}_behive_like_profile_normed.pdf\", bbox_inches = 'tight')"
+    "if save_fig: plt.savefig(f\"{output_prefix}_behive_like_profile_normed.pdf\", bbox_inches = 'tight')"
    ]
   },
   {
@@ -1020,7 +1021,8 @@
    "source": [
     "if pam_col is not None:\n",
     "    pos_by_pam = be.pl.editing_patterns.plot_by_pos_pam(cdata_bulk, cedit_rates_df_ag, pam_col)\n",
-    "    plt.savefig(f\"{output_prefix}_pos_by_pam.pdf\", bbox_inches = 'tight')\n",
+    "    if save_fig: \n",
+    "        plt.savefig(f\"{output_prefix}_pos_by_pam.pdf\", bbox_inches = 'tight')\n",
     "    pos_by_pam.to_csv(f\"{output_prefix}_pos_by_pam.csv\")"
    ]
   },
@@ -1068,9 +1070,14 @@
    "outputs": [],
    "source": [
     "con_mean_er = {}\n",
+    "bases = [\"A\", \"C\", \"G\", \"T\"]\n",
+    "\n",
     "for i in range(3):\n",
+    "\n",
     "    cedit_rates_df_ag_window[f\"context_{i}\"] = cedit_rates_df_ag_window.context.map(lambda s: s[i])\n",
-    "    con_mean_er[i] = cedit_rates_df_ag_window.groupby(f\"context_{i}\")[\"rep_mean\"].mean()"
+    "\n",
+    "    con_mean_er[i] = cedit_rates_df_ag_window.groupby(f\"context_{i}\")[\"rep_mean\"].mean()\n",
+    "    con_mean_er[i] = con_mean_er[i].reindex(bases).fillna(0)"
    ]
   },
   {
@@ -1115,7 +1122,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "bases = [\"A\", \"C\", \"G\", \"T\"]\n",
+    "\n",
     "target_df = pd.DataFrame([int(b == cdata_bulk.uns[\"target_base_change\"][0]) for b in bases], index=bases)\n",
     "ic_tbl = pd.concat(\n",
     "    [\n",
@@ -1138,13 +1145,6 @@
     "ic_tbl.to_csv(f\"{output_prefix}_context.csv\")"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  },
   {
    "cell_type": "code",
    "execution_count": 32,
@@ -1180,7 +1180,7 @@
     "\n",
     "logomaker.Logo(ic_tbl, ax = ax)\n",
     "ax.set_ylabel(\"Relative frequency\")\n",
-    "fig.savefig(f\"{output_prefix}_context_preference_{window_start}_{window_end}.pdf\")"
+    "if save_fig: fig.savefig(f\"{output_prefix}_context_preference_{window_start}_{window_end}.pdf\")"
    ]
   }
  ],

bean/plotting/utils.py

@@ -45,13 +45,22 @@ def parse_args(parser=None):
         type=int,
         default=6,
     )
+    parser.add_argument(
+        "--save-fig",
+        action="store_true",
+        help="Save .pdf of the figures included in the report.",
+    )
 
     return parser
 
 
 def check_args(args):
     if args.output_prefix is None:
-        args.output_prefix = os.path.splitext(args.bdata_path)[0]
+        sample_id = os.path.splitext(os.path.basename(args.bdata_path))[0]
+        args.output_prefix = (
+            f"{os.path.dirname(args.bdata_path)}/bean_profile.{sample_id}/{sample_id}"
+        )
+        os.makedirs(args.output_prefix, exist_ok=True)
     if args.window_length < 1:
         raise ValueError(f"window_length {args.window_length} is too small.")
     if args.window_length > 20: