Dynamically make list of flu clades to plot

Instead of hardcoding a list of flu clades to plot in early and late
data, build the list of clades dynamically from the input table. For the
later flu data, only assign colors to clades with at least 10 samples,
so we can more easily distinguish these larger groups and not waste the
few colors we have on smaller groups.

seasonal-flu-nextstrain-2018-2020/2021-03-09Notebook.ipynb

@@ -67,140 +67,164 @@
     "static_mds_chart = snakemake.output.MDS_Supplement_PNG"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "dc772cba",
+   "metadata": {},
+   "source": [
+    "## Load data"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "5e135bbc",
+   "id": "06532877",
    "metadata": {},
    "outputs": [],
    "source": [
-    "clades_to_plot = [\n",
-    "    '3c2',\n",
-    "    '3c3.A',\n",
-    "    'A1',\n",
-    "    'A1b/131K',\n",
-    "    'A1b/135K',\n",
-    "    'A1b/135N',\n",
-    "    'A1b/137F',\n",
-    "    'A1b/186D',\n",
-    "    'A1b/197R',\n",
-    "    'A1b/94N',\n",
-    "    'A2',\n",
-    "    'A3',\n",
-    "]\n",
-    "domain = clades_to_plot"
+    "embeddings_df = pd.read_csv(embeddings_path, sep=\"\\t\")"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "51de9a8a",
+   "id": "bc994d18",
    "metadata": {},
    "outputs": [],
    "source": [
-    "len(clades_to_plot)"
+    "embeddings_df = embeddings_df.rename(\n",
+    "    columns={\n",
+    "        \"num_date\": \"date\",\n",
+    "    }\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "53923a14",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "clade_counts = embeddings_df[\"clade_membership\"].value_counts()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "2b3a1e7b",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "clade_counts"
    ]
   },
   {
    "cell_type": "markdown",
-   "id": "dc772cba",
+   "id": "f10ad3f9",
    "metadata": {},
    "source": [
-    "## Load data"
+    "Only assign colors to clades with at least 10 samples. This approach allows us to clearly see larger clades using fewer colors."
    ]
   },
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "3b0b80a2",
+   "id": "5e4c5882",
    "metadata": {},
    "outputs": [],
    "source": [
-    "colors = pd.read_csv(colors_path, sep=\"\\t\", names=[i for i in range(0,101)], nrows=101)"
+    "clades_to_plot_with_color = sorted(clade_counts[clade_counts >= 10].index.values)"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "28c2742c",
+   "id": "e4570623",
    "metadata": {},
    "outputs": [],
    "source": [
-    "clade_color_range = colors.iloc[len(clades_to_plot) - 1].dropna().tolist()"
+    "clades_to_plot = sorted(embeddings_df[\"clade_membership\"].drop_duplicates().values)"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "57ada75a",
+   "id": "de1a59a2",
    "metadata": {},
    "outputs": [],
    "source": [
-    "len(clade_color_range)"
+    "clades_to_plot"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "40527d62",
+   "id": "5e135bbc",
    "metadata": {},
    "outputs": [],
    "source": [
-    "#domain.append(\"other\")"
+    "domain = clades_to_plot_with_color"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "75fb5a9d",
+   "id": "51de9a8a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "len(clades_to_plot)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "3b0b80a2",
    "metadata": {},
    "outputs": [],
    "source": [
-    "#clade_color_range.append(\"#999999\")"
+    "colors = pd.read_csv(colors_path, sep=\"\\t\", names=[i for i in range(0,101)], nrows=101)"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "06532877",
+   "id": "28c2742c",
    "metadata": {},
    "outputs": [],
    "source": [
-    "embeddings_df = pd.read_csv(embeddings_path, sep=\"\\t\")"
+    "clade_color_range = colors.iloc[len(clades_to_plot_with_color) - 1].dropna().tolist()"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "bc994d18",
+   "id": "57ada75a",
    "metadata": {},
    "outputs": [],
    "source": [
-    "embeddings_df = embeddings_df.rename(\n",
-    "    columns={\n",
-    "        \"num_date\": \"date\",\n",
-    "    }\n",
-    ")"
+    "len(clade_color_range)"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "53923a14",
+   "id": "40527d62",
    "metadata": {},
    "outputs": [],
    "source": [
-    "embeddings_df[\"clade_membership\"].value_counts()"
+    "domain.append(\"other\")"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "e4570623",
+   "id": "75fb5a9d",
    "metadata": {},
    "outputs": [],
    "source": [
-    "embeddings_df[\"clade_membership\"].drop_duplicates().sort_values().tolist()"
+    "clade_color_range.append(\"#999999\")"
    ]
   },
   {
@@ -211,7 +235,7 @@
    "outputs": [],
    "source": [
     "embeddings_df[\"clade_membership_color\"] = embeddings_df[\"clade_membership\"].apply(\n",
-    "    lambda clade: clade if clade in clades_to_plot else \"other\"\n",
+    "    lambda clade: clade if clade in clades_to_plot_with_color else \"other\"\n",
     ")"
    ]
   },
@@ -225,6 +249,36 @@
     "embeddings_df.head()"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "dfafbe8e",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "embeddings_df[\"clade_membership_color\"].value_counts()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "b6652ac0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "domain"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "eba1af54",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "clade_color_range"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,