{ "cells": [ { "cell_type": "markdown", "id": "53215a88-f707-47c1-8a59-76118e9180f8", "metadata": {}, "source": [ "# Analysis of differences with different gene to disease sources" ] }, { "cell_type": "markdown", "id": "49a45dcd-45ce-4ddf-9d40-a9854eff5bc6", "metadata": {}, "source": [ "## Load all associations from multiple sources\n", "\n", "The mondo-g2d input spec includes metadata about where to download\n", "\n", "- gencc\n", "- medgen_mim2gene\n", "\n", "Note these use heterogeneous IDs, the spec includes normalizers for these" ] }, { "cell_type": "code", "execution_count": 1, "id": "14c7fa8d", "metadata": {}, "outputs": [], "source": [ "from oaklib import get_adapter" ] }, { "cell_type": "code", "execution_count": 2, "id": "398637c9", "metadata": {}, "outputs": [], "source": [ "from oaklib.conf import CONF_DIR_PATH\n", "adapter = get_adapter(CONF_DIR_PATH/ \"mondo-g2d-input-spec.yaml\")" ] }, { "cell_type": "code", "execution_count": 3, "id": "bb0d2772", "metadata": {}, "outputs": [], "source": [ "associations = list(adapter.associations())" ] }, { "cell_type": "code", "execution_count": 4, "id": "7010824d", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "106577" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "len(associations)" ] }, { "cell_type": "code", "execution_count": 5, "id": "0f04152e", "metadata": {}, "outputs": [], "source": [ "from linkml_runtime.dumpers import json_dumper" ] }, { "cell_type": "code", "execution_count": 6, "id": "249f2862", "metadata": {}, "outputs": [], "source": [ "import pandas as pd" ] }, { "cell_type": "code", "execution_count": 7, "id": "e0ccb776-a781-466d-9d04-073bb3b41a96", "metadata": {}, "outputs": [], "source": [ "def _as_dict(a):\n", " obj = json_dumper.to_dict(a)\n", " obj[\"subject_prefix\"] = a.subject.split(\":\")[0]\n", " obj[\"object_prefix\"] = a.object.split(\":\")[0]\n", " return obj" ] }, { "cell_type": "code", "execution_count": 8, "id": "17d15e7e", "metadata": {}, "outputs": [], "source": [ "df = pd.DataFrame([_as_dict(a) for a in associations])\n", "df.to_csv(\"g2d.tsv\", sep=\"\\t\", index=False)" ] }, { "cell_type": "code", "execution_count": 9, "id": "41f88348", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
subjectpredicateobjectprimary_knowledge_sourceaggregator_knowledge_sourceoriginal_objectoriginal_subjectsubject_prefixobject_prefixsubject_labelobject_label
0HGNC:1952biolink:gene_associated_with_conditionMONDO:0007032infores:medgen_mim_g2dinfores:medgen_mim_g2dOMIM:100100NCBIGene:1131HGNCMONDONaNNaN
1HGNC:29216biolink:gene_associated_with_conditionMONDO:0024506infores:medgen_mim_g2dinfores:medgen_mim_g2dOMIM:100300NCBIGene:57514HGNCMONDONaNNaN
2HGNC:3690biolink:gene_associated_with_conditionMONDO:0007037infores:medgen_mim_g2dinfores:medgen_mim_g2dOMIM:100800NCBIGene:2261HGNCMONDONaNNaN
3HGNC:7773biolink:gene_associated_with_conditionMONDO:0007039infores:medgen_mim_g2dinfores:medgen_mim_g2dOMIM:101000NCBIGene:4771HGNCMONDONaNNaN
4HGNC:3689biolink:gene_associated_with_conditionMONDO:0007041infores:medgen_mim_g2dinfores:medgen_mim_g2dOMIM:101200NCBIGene:2263HGNCMONDONaNNaN
....................................
106572HGNC:12762NaNMONDO:0013673GENCC:000110infores:genccNaNNaNHGNCMONDOWFS1Wolfram-like syndrome
106573HGNC:12762NaNMONDO:0013673GENCC:000102infores:genccNaNNaNHGNCMONDOWFS1Wolfram-like syndrome
106574HGNC:12762NaNMONDO:0013673GENCC:000101infores:genccNaNNaNHGNCMONDOWFS1Wolfram-like syndrome
106575HGNC:29271NaNMONDO:0030491ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_...infores:hpoaOMIM:619644NCBIGene:57169HGNCMONDOZNFX1NaN
106576HGNC:29271NaNMONDO:0030491GENCC:000101infores:genccNaNNaNHGNCMONDOZNFX1immunodeficiency 91 and hyperinflammation
\n", "

106577 rows × 11 columns

\n", "
" ], "text/plain": [ " subject predicate object \\\n", "0 HGNC:1952 biolink:gene_associated_with_condition MONDO:0007032 \n", "1 HGNC:29216 biolink:gene_associated_with_condition MONDO:0024506 \n", "2 HGNC:3690 biolink:gene_associated_with_condition MONDO:0007037 \n", "3 HGNC:7773 biolink:gene_associated_with_condition MONDO:0007039 \n", "4 HGNC:3689 biolink:gene_associated_with_condition MONDO:0007041 \n", "... ... ... ... \n", "106572 HGNC:12762 NaN MONDO:0013673 \n", "106573 HGNC:12762 NaN MONDO:0013673 \n", "106574 HGNC:12762 NaN MONDO:0013673 \n", "106575 HGNC:29271 NaN MONDO:0030491 \n", "106576 HGNC:29271 NaN MONDO:0030491 \n", "\n", " primary_knowledge_source \\\n", "0 infores:medgen_mim_g2d \n", "1 infores:medgen_mim_g2d \n", "2 infores:medgen_mim_g2d \n", "3 infores:medgen_mim_g2d \n", "4 infores:medgen_mim_g2d \n", "... ... \n", "106572 GENCC:000110 \n", "106573 GENCC:000102 \n", "106574 GENCC:000101 \n", "106575 ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_... \n", "106576 GENCC:000101 \n", "\n", " aggregator_knowledge_source original_object original_subject \\\n", "0 infores:medgen_mim_g2d OMIM:100100 NCBIGene:1131 \n", "1 infores:medgen_mim_g2d OMIM:100300 NCBIGene:57514 \n", "2 infores:medgen_mim_g2d OMIM:100800 NCBIGene:2261 \n", "3 infores:medgen_mim_g2d OMIM:101000 NCBIGene:4771 \n", "4 infores:medgen_mim_g2d OMIM:101200 NCBIGene:2263 \n", "... ... ... ... \n", "106572 infores:gencc NaN NaN \n", "106573 infores:gencc NaN NaN \n", "106574 infores:gencc NaN NaN \n", "106575 infores:hpoa OMIM:619644 NCBIGene:57169 \n", "106576 infores:gencc NaN NaN \n", "\n", " subject_prefix object_prefix subject_label \\\n", "0 HGNC MONDO NaN \n", "1 HGNC MONDO NaN \n", "2 HGNC MONDO NaN \n", "3 HGNC MONDO NaN \n", "4 HGNC MONDO NaN \n", "... ... ... ... \n", "106572 HGNC MONDO WFS1 \n", "106573 HGNC MONDO WFS1 \n", "106574 HGNC MONDO WFS1 \n", "106575 HGNC MONDO ZNFX1 \n", "106576 HGNC MONDO ZNFX1 \n", "\n", " object_label \n", "0 NaN \n", "1 NaN \n", "2 NaN \n", "3 NaN \n", "4 NaN \n", "... ... \n", "106572 Wolfram-like syndrome \n", "106573 Wolfram-like syndrome \n", "106574 Wolfram-like syndrome \n", "106575 NaN \n", "106576 immunodeficiency 91 and hyperinflammation \n", "\n", "[106577 rows x 11 columns]" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df" ] }, { "cell_type": "code", "execution_count": 10, "id": "c24ada89", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
subjectpredicateobjectprimary_knowledge_sourceaggregator_knowledge_sourceoriginal_objectoriginal_subjectsubject_prefixobject_prefixsubject_labelobject_label
699HGNC:10896biolink:gene_associated_with_conditionMONDO:0008426infores:medgen_mim_g2dinfores:medgen_mim_g2dOMIM:182212NCBIGene:6497HGNCMONDONaNNaN
15644HGNC:10896NaNMONDO:0008426ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_...infores:hpoaOMIM:182212NCBIGene:6497HGNCMONDOSKINaN
15645HGNC:10896NaNMONDO:0008426http://www.orphadata.org/data/xml/en_product6.xmlinfores:hpoaOrphanet:2462NCBIGene:6497HGNCMONDOSKINaN
15646HGNC:10896NaNMONDO:0008426GENCC:000101infores:genccNaNNaNHGNCMONDOSKIShprintzen-Goldberg syndrome
15647HGNC:10896NaNMONDO:0008426GENCC:000112infores:genccNaNNaNHGNCMONDOSKIShprintzen-Goldberg syndrome
....................................
104537HGNC:10896NaNMONDO:0008426GENCC:000111infores:genccNaNNaNHGNCMONDOSKIShprintzen-Goldberg syndrome
104538HGNC:10896NaNMONDO:0008426GENCC:000104infores:genccNaNNaNHGNCMONDOSKIShprintzen-Goldberg syndrome
104539HGNC:10896NaNMONDO:0008426GENCC:000106infores:genccNaNNaNHGNCMONDOSKIShprintzen-Goldberg syndrome
104540HGNC:10896NaNMONDO:0008426GENCC:000110infores:genccNaNNaNHGNCMONDOSKIShprintzen-Goldberg syndrome
104541HGNC:10896NaNMONDO:0008426GENCC:000102infores:genccNaNNaNHGNCMONDOSKIShprintzen-Goldberg syndrome
\n", "

98 rows × 11 columns

\n", "
" ], "text/plain": [ " subject predicate object \\\n", "699 HGNC:10896 biolink:gene_associated_with_condition MONDO:0008426 \n", "15644 HGNC:10896 NaN MONDO:0008426 \n", "15645 HGNC:10896 NaN MONDO:0008426 \n", "15646 HGNC:10896 NaN MONDO:0008426 \n", "15647 HGNC:10896 NaN MONDO:0008426 \n", "... ... ... ... \n", "104537 HGNC:10896 NaN MONDO:0008426 \n", "104538 HGNC:10896 NaN MONDO:0008426 \n", "104539 HGNC:10896 NaN MONDO:0008426 \n", "104540 HGNC:10896 NaN MONDO:0008426 \n", "104541 HGNC:10896 NaN MONDO:0008426 \n", "\n", " primary_knowledge_source \\\n", "699 infores:medgen_mim_g2d \n", "15644 ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_... \n", "15645 http://www.orphadata.org/data/xml/en_product6.xml \n", "15646 GENCC:000101 \n", "15647 GENCC:000112 \n", "... ... \n", "104537 GENCC:000111 \n", "104538 GENCC:000104 \n", "104539 GENCC:000106 \n", "104540 GENCC:000110 \n", "104541 GENCC:000102 \n", "\n", " aggregator_knowledge_source original_object original_subject \\\n", "699 infores:medgen_mim_g2d OMIM:182212 NCBIGene:6497 \n", "15644 infores:hpoa OMIM:182212 NCBIGene:6497 \n", "15645 infores:hpoa Orphanet:2462 NCBIGene:6497 \n", "15646 infores:gencc NaN NaN \n", "15647 infores:gencc NaN NaN \n", "... ... ... ... \n", "104537 infores:gencc NaN NaN \n", "104538 infores:gencc NaN NaN \n", "104539 infores:gencc NaN NaN \n", "104540 infores:gencc NaN NaN \n", "104541 infores:gencc NaN NaN \n", "\n", " subject_prefix object_prefix subject_label \\\n", "699 HGNC MONDO NaN \n", "15644 HGNC MONDO SKI \n", "15645 HGNC MONDO SKI \n", "15646 HGNC MONDO SKI \n", "15647 HGNC MONDO SKI \n", "... ... ... ... \n", "104537 HGNC MONDO SKI \n", "104538 HGNC MONDO SKI \n", "104539 HGNC MONDO SKI \n", "104540 HGNC MONDO SKI \n", "104541 HGNC MONDO SKI \n", "\n", " object_label \n", "699 NaN \n", "15644 NaN \n", "15645 NaN \n", "15646 Shprintzen-Goldberg syndrome \n", "15647 Shprintzen-Goldberg syndrome \n", "... ... \n", "104537 Shprintzen-Goldberg syndrome \n", "104538 Shprintzen-Goldberg syndrome \n", "104539 Shprintzen-Goldberg syndrome \n", "104540 Shprintzen-Goldberg syndrome \n", "104541 Shprintzen-Goldberg syndrome \n", "\n", "[98 rows x 11 columns]" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.query(\"object == 'MONDO:0008426'\")" ] }, { "cell_type": "code", "execution_count": 11, "id": "53a860e5-4162-4456-915e-6c18a6bbeb45", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
subjectpredicateobjectprimary_knowledge_sourceaggregator_knowledge_sourceoriginal_objectoriginal_subjectsubject_prefixobject_prefixsubject_labelobject_label
5907HGNC:16391NaNMONDO:0008905ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_...infores:hpoaOMIM:212050NCBIGene:64170HGNCMONDOCARD9NaN
5908HGNC:16391NaNMONDO:0008905http://www.orphadata.org/data/xml/en_product6.xmlinfores:hpoaOrphanet:457088NCBIGene:64170HGNCMONDOCARD9NaN
5911HGNC:21066NaNMONDO:0009544ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_...infores:hpoaOMIM:248000NCBIGene:51256HGNCMONDOTBC1D7NaN
5915HGNC:14313NaNMONDO:0033485ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_...infores:hpoaOMIM:617895NCBIGene:28981HGNCMONDOIFT81NaN
5916HGNC:6742NaNMONDO:0014693ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_...infores:hpoaOMIM:616564NCBIGene:8216HGNCMONDOLZTR1NaN
....................................
106562HGNC:12731NaNMONDO:0010294ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_...infores:hpoaOMIM:300299NCBIGene:7454HGNCMONDOWASNaN
106563HGNC:12731NaNMONDO:0010294http://www.orphadata.org/data/xml/en_product6.xmlinfores:hpoaOrphanet:86788NCBIGene:7454HGNCMONDOWASNaN
106567HGNC:12762NaNMONDO:0013673ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_...infores:hpoaOMIM:614296NCBIGene:7466HGNCMONDOWFS1NaN
106568HGNC:12762NaNMONDO:0013673http://www.orphadata.org/data/xml/en_product6.xmlinfores:hpoaOrphanet:411590NCBIGene:7466HGNCMONDOWFS1NaN
106575HGNC:29271NaNMONDO:0030491ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_...infores:hpoaOMIM:619644NCBIGene:57169HGNCMONDOZNFX1NaN
\n", "

40613 rows × 11 columns

\n", "
" ], "text/plain": [ " subject predicate object \\\n", "5907 HGNC:16391 NaN MONDO:0008905 \n", "5908 HGNC:16391 NaN MONDO:0008905 \n", "5911 HGNC:21066 NaN MONDO:0009544 \n", "5915 HGNC:14313 NaN MONDO:0033485 \n", "5916 HGNC:6742 NaN MONDO:0014693 \n", "... ... ... ... \n", "106562 HGNC:12731 NaN MONDO:0010294 \n", "106563 HGNC:12731 NaN MONDO:0010294 \n", "106567 HGNC:12762 NaN MONDO:0013673 \n", "106568 HGNC:12762 NaN MONDO:0013673 \n", "106575 HGNC:29271 NaN MONDO:0030491 \n", "\n", " primary_knowledge_source \\\n", "5907 ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_... \n", "5908 http://www.orphadata.org/data/xml/en_product6.xml \n", "5911 ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_... \n", "5915 ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_... \n", "5916 ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_... \n", "... ... \n", "106562 ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_... \n", "106563 http://www.orphadata.org/data/xml/en_product6.xml \n", "106567 ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_... \n", "106568 http://www.orphadata.org/data/xml/en_product6.xml \n", "106575 ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_... \n", "\n", " aggregator_knowledge_source original_object original_subject \\\n", "5907 infores:hpoa OMIM:212050 NCBIGene:64170 \n", "5908 infores:hpoa Orphanet:457088 NCBIGene:64170 \n", "5911 infores:hpoa OMIM:248000 NCBIGene:51256 \n", "5915 infores:hpoa OMIM:617895 NCBIGene:28981 \n", "5916 infores:hpoa OMIM:616564 NCBIGene:8216 \n", "... ... ... ... \n", "106562 infores:hpoa OMIM:300299 NCBIGene:7454 \n", "106563 infores:hpoa Orphanet:86788 NCBIGene:7454 \n", "106567 infores:hpoa OMIM:614296 NCBIGene:7466 \n", "106568 infores:hpoa Orphanet:411590 NCBIGene:7466 \n", "106575 infores:hpoa OMIM:619644 NCBIGene:57169 \n", "\n", " subject_prefix object_prefix subject_label object_label \n", "5907 HGNC MONDO CARD9 NaN \n", "5908 HGNC MONDO CARD9 NaN \n", "5911 HGNC MONDO TBC1D7 NaN \n", "5915 HGNC MONDO IFT81 NaN \n", "5916 HGNC MONDO LZTR1 NaN \n", "... ... ... ... ... \n", "106562 HGNC MONDO WAS NaN \n", "106563 HGNC MONDO WAS NaN \n", "106567 HGNC MONDO WFS1 NaN \n", "106568 HGNC MONDO WFS1 NaN \n", "106575 HGNC MONDO ZNFX1 NaN \n", "\n", "[40613 rows x 11 columns]" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.query(\"aggregator_knowledge_source == 'infores:hpoa'\")" ] }, { "cell_type": "code", "execution_count": 12, "id": "7465ff08", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
countuniquetopfreq
subjectobject
HGNC:10001MONDO:000778142infores:hpoa2
HGNC:10004MONDO:0012033503infores:gencc35
HGNC:10006MONDO:0008493173infores:hpoa8
MONDO:0019107103infores:hpoa6
HGNC:10008MONDO:001910742infores:hpoa2
..................
NCBIGene:105259599MONDO:002079622infores:medgen_mim_g2d1
NCBIGene:109580095MONDO:001351722infores:medgen_mim_g2d1
NCBIGene:111365204MONDO:000763022infores:medgen_mim_g2d1
MONDO:001093222infores:medgen_mim_g2d1
NCBIGene:7467MONDO:000868422infores:medgen_mim_g2d1
\n", "

15696 rows × 4 columns

\n", "
" ], "text/plain": [ " count unique top freq\n", "subject object \n", "HGNC:10001 MONDO:0007781 4 2 infores:hpoa 2\n", "HGNC:10004 MONDO:0012033 50 3 infores:gencc 35\n", "HGNC:10006 MONDO:0008493 17 3 infores:hpoa 8\n", " MONDO:0019107 10 3 infores:hpoa 6\n", "HGNC:10008 MONDO:0019107 4 2 infores:hpoa 2\n", "... ... ... ... ...\n", "NCBIGene:105259599 MONDO:0020796 2 2 infores:medgen_mim_g2d 1\n", "NCBIGene:109580095 MONDO:0013517 2 2 infores:medgen_mim_g2d 1\n", "NCBIGene:111365204 MONDO:0007630 2 2 infores:medgen_mim_g2d 1\n", " MONDO:0010932 2 2 infores:medgen_mim_g2d 1\n", "NCBIGene:7467 MONDO:0008684 2 2 infores:medgen_mim_g2d 1\n", "\n", "[15696 rows x 4 columns]" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.groupby([\"subject\", \"object\"])[\"aggregator_knowledge_source\"].describe()" ] }, { "cell_type": "code", "execution_count": 13, "id": "fce47a73-4d81-423e-ab32-72b68752e9e1", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
countuniquetopfreq
subject_prefixobject_prefix
HGNCMONDO1057103infores:gencc60057
OMIM7192infores:hpoa405
Orphanet1321infores:hpoa132
NCBIGeneMONDO162infores:medgen_mim_g2d8
\n", "
" ], "text/plain": [ " count unique top freq\n", "subject_prefix object_prefix \n", "HGNC MONDO 105710 3 infores:gencc 60057\n", " OMIM 719 2 infores:hpoa 405\n", " Orphanet 132 1 infores:hpoa 132\n", "NCBIGene MONDO 16 2 infores:medgen_mim_g2d 8" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.groupby([\"subject_prefix\", \"object_prefix\"])[\"aggregator_knowledge_source\"].describe()" ] }, { "cell_type": "code", "execution_count": 14, "id": "fac1aa12-e69e-4406-9f66-a0633018d660", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
subjectpredicateobject...original_subjectsubject_labelobject_label
countuniquetopfreqcountuniquetopfreqcountunique...topfreqcountuniquetopfreqcountuniquetopfreq
subject_prefixobject_prefixaggregator_knowledge_source
HGNCMONDOinfores:gencc600574753HGNC:369024600NaNNaN600576155...NaNNaN600574753FGFR3246600576151retinitis pigmentosa229
infores:hpoa400685128HGNC:220013000NaNNaN400687527...NCBIGene:1280130400685128COL2A113000NaNNaN
infores:medgen_mim_g2d55853933HGNC:22001455851biolink:gene_associated_with_condition558555855093...NCBIGene:12801400NaNNaN00NaNNaN
OMIMinfores:hpoa405385HGNC:4398200NaNNaN405404...NCBIGene:27832405385GNB2200NaNNaN
infores:medgen_mim_g2d314305HGNC:1842023141biolink:gene_associated_with_condition314314314...NCBIGene:29072200NaNNaN00NaNNaN
Orphanetinfores:hpoa132111HGNC:7577200NaNNaN13283...NCBIGene:46252132111MYH7200NaNNaN
NCBIGeneMONDOinfores:hpoa85NCBIGene:105259599300NaNNaN88...NaNNaN81-800NaNNaN
infores:medgen_mim_g2d85NCBIGene:105259599381biolink:gene_associated_with_condition888...NaNNaN00NaNNaN00NaNNaN
\n", "

8 rows × 32 columns

\n", "
" ], "text/plain": [ " subject \\\n", " count unique \n", "subject_prefix object_prefix aggregator_knowledge_source \n", "HGNC MONDO infores:gencc 60057 4753 \n", " infores:hpoa 40068 5128 \n", " infores:medgen_mim_g2d 5585 3933 \n", " OMIM infores:hpoa 405 385 \n", " infores:medgen_mim_g2d 314 305 \n", " Orphanet infores:hpoa 132 111 \n", "NCBIGene MONDO infores:hpoa 8 5 \n", " infores:medgen_mim_g2d 8 5 \n", "\n", " \\\n", " top \n", "subject_prefix object_prefix aggregator_knowledge_source \n", "HGNC MONDO infores:gencc HGNC:3690 \n", " infores:hpoa HGNC:2200 \n", " infores:medgen_mim_g2d HGNC:2200 \n", " OMIM infores:hpoa HGNC:4398 \n", " infores:medgen_mim_g2d HGNC:18420 \n", " Orphanet infores:hpoa HGNC:7577 \n", "NCBIGene MONDO infores:hpoa NCBIGene:105259599 \n", " infores:medgen_mim_g2d NCBIGene:105259599 \n", "\n", " predicate \\\n", " freq count \n", "subject_prefix object_prefix aggregator_knowledge_source \n", "HGNC MONDO infores:gencc 246 0 \n", " infores:hpoa 130 0 \n", " infores:medgen_mim_g2d 14 5585 \n", " OMIM infores:hpoa 2 0 \n", " infores:medgen_mim_g2d 2 314 \n", " Orphanet infores:hpoa 2 0 \n", "NCBIGene MONDO infores:hpoa 3 0 \n", " infores:medgen_mim_g2d 3 8 \n", "\n", " \\\n", " unique \n", "subject_prefix object_prefix aggregator_knowledge_source \n", "HGNC MONDO infores:gencc 0 \n", " infores:hpoa 0 \n", " infores:medgen_mim_g2d 1 \n", " OMIM infores:hpoa 0 \n", " infores:medgen_mim_g2d 1 \n", " Orphanet infores:hpoa 0 \n", "NCBIGene MONDO infores:hpoa 0 \n", " infores:medgen_mim_g2d 1 \n", "\n", " \\\n", " top \n", "subject_prefix object_prefix aggregator_knowledge_source \n", "HGNC MONDO infores:gencc NaN \n", " infores:hpoa NaN \n", " infores:medgen_mim_g2d biolink:gene_associated_with_condition \n", " OMIM infores:hpoa NaN \n", " infores:medgen_mim_g2d biolink:gene_associated_with_condition \n", " Orphanet infores:hpoa NaN \n", "NCBIGene MONDO infores:hpoa NaN \n", " infores:medgen_mim_g2d biolink:gene_associated_with_condition \n", "\n", " object \\\n", " freq count unique \n", "subject_prefix object_prefix aggregator_knowledge_source \n", "HGNC MONDO infores:gencc NaN 60057 6155 \n", " infores:hpoa NaN 40068 7527 \n", " infores:medgen_mim_g2d 5585 5585 5093 \n", " OMIM infores:hpoa NaN 405 404 \n", " infores:medgen_mim_g2d 314 314 314 \n", " Orphanet infores:hpoa NaN 132 83 \n", "NCBIGene MONDO infores:hpoa NaN 8 8 \n", " infores:medgen_mim_g2d 8 8 8 \n", "\n", " ... \\\n", " ... \n", "subject_prefix object_prefix aggregator_knowledge_source ... \n", "HGNC MONDO infores:gencc ... \n", " infores:hpoa ... \n", " infores:medgen_mim_g2d ... \n", " OMIM infores:hpoa ... \n", " infores:medgen_mim_g2d ... \n", " Orphanet infores:hpoa ... \n", "NCBIGene MONDO infores:hpoa ... \n", " infores:medgen_mim_g2d ... \n", "\n", " original_subject \\\n", " top \n", "subject_prefix object_prefix aggregator_knowledge_source \n", "HGNC MONDO infores:gencc NaN \n", " infores:hpoa NCBIGene:1280 \n", " infores:medgen_mim_g2d NCBIGene:1280 \n", " OMIM infores:hpoa NCBIGene:2783 \n", " infores:medgen_mim_g2d NCBIGene:29072 \n", " Orphanet infores:hpoa NCBIGene:4625 \n", "NCBIGene MONDO infores:hpoa NaN \n", " infores:medgen_mim_g2d NaN \n", "\n", " subject_label \\\n", " freq count \n", "subject_prefix object_prefix aggregator_knowledge_source \n", "HGNC MONDO infores:gencc NaN 60057 \n", " infores:hpoa 130 40068 \n", " infores:medgen_mim_g2d 14 0 \n", " OMIM infores:hpoa 2 405 \n", " infores:medgen_mim_g2d 2 0 \n", " Orphanet infores:hpoa 2 132 \n", "NCBIGene MONDO infores:hpoa NaN 8 \n", " infores:medgen_mim_g2d NaN 0 \n", "\n", " \\\n", " unique top freq \n", "subject_prefix object_prefix aggregator_knowledge_source \n", "HGNC MONDO infores:gencc 4753 FGFR3 246 \n", " infores:hpoa 5128 COL2A1 130 \n", " infores:medgen_mim_g2d 0 NaN NaN \n", " OMIM infores:hpoa 385 GNB2 2 \n", " infores:medgen_mim_g2d 0 NaN NaN \n", " Orphanet infores:hpoa 111 MYH7 2 \n", "NCBIGene MONDO infores:hpoa 1 - 8 \n", " infores:medgen_mim_g2d 0 NaN NaN \n", "\n", " object_label \\\n", " count unique \n", "subject_prefix object_prefix aggregator_knowledge_source \n", "HGNC MONDO infores:gencc 60057 6151 \n", " infores:hpoa 0 0 \n", " infores:medgen_mim_g2d 0 0 \n", " OMIM infores:hpoa 0 0 \n", " infores:medgen_mim_g2d 0 0 \n", " Orphanet infores:hpoa 0 0 \n", "NCBIGene MONDO infores:hpoa 0 0 \n", " infores:medgen_mim_g2d 0 0 \n", "\n", " \\\n", " top \n", "subject_prefix object_prefix aggregator_knowledge_source \n", "HGNC MONDO infores:gencc retinitis pigmentosa \n", " infores:hpoa NaN \n", " infores:medgen_mim_g2d NaN \n", " OMIM infores:hpoa NaN \n", " infores:medgen_mim_g2d NaN \n", " Orphanet infores:hpoa NaN \n", "NCBIGene MONDO infores:hpoa NaN \n", " infores:medgen_mim_g2d NaN \n", "\n", " \n", " freq \n", "subject_prefix object_prefix aggregator_knowledge_source \n", "HGNC MONDO infores:gencc 229 \n", " infores:hpoa NaN \n", " infores:medgen_mim_g2d NaN \n", " OMIM infores:hpoa NaN \n", " infores:medgen_mim_g2d NaN \n", " Orphanet infores:hpoa NaN \n", "NCBIGene MONDO infores:hpoa NaN \n", " infores:medgen_mim_g2d NaN \n", "\n", "[8 rows x 32 columns]" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.groupby([\"subject_prefix\", \"object_prefix\", \"aggregator_knowledge_source\"]).describe()" ] }, { "cell_type": "code", "execution_count": 15, "id": "a7919a4b", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
subjectpredicateobjectprimary_knowledge_sourceaggregator_knowledge_sourceoriginal_objectoriginal_subjectsubject_prefixobject_prefixsubject_labelobject_label
16423HGNC:10001NaNMONDO:0007781ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_...infores:hpoaOMIM:145500NCBIGene:8490HGNCMONDORGS5NaN
16424HGNC:10001NaNMONDO:0007781GENCC:000106infores:genccNaNNaNHGNCMONDORGS5essential hypertension, genetic
47487HGNC:10001NaNMONDO:0007781ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_...infores:hpoaOMIM:145500NCBIGene:8490HGNCMONDORGS5NaN
47488HGNC:10001NaNMONDO:0007781GENCC:000106infores:genccNaNNaNHGNCMONDORGS5essential hypertension, genetic
2684HGNC:10004biolink:gene_associated_with_conditionMONDO:0012033infores:medgen_mim_g2dinfores:medgen_mim_g2dOMIM:608415NCBIGene:8787HGNCMONDONaNNaN
....................................
7192NCBIGene:111365204NaNMONDO:0007630ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_...infores:hpoaOMIM:136550NaNNCBIGeneMONDO-NaN
2108NCBIGene:111365204biolink:gene_associated_with_conditionMONDO:0010932infores:medgen_mim_g2dinfores:medgen_mim_g2dOMIM:600790NaNNCBIGeneMONDONaNNaN
9063NCBIGene:111365204NaNMONDO:0010932ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_...infores:hpoaOMIM:600790NaNNCBIGeneMONDO-NaN
817NCBIGene:7467biolink:gene_associated_with_conditionMONDO:0008684infores:medgen_mim_g2dinfores:medgen_mim_g2dOMIM:194190NaNNCBIGeneMONDONaNNaN
11624NCBIGene:7467NaNMONDO:0008684ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_...infores:hpoaOMIM:194190NaNNCBIGeneMONDO-NaN
\n", "

106577 rows × 11 columns

\n", "
" ], "text/plain": [ " subject predicate \\\n", "16423 HGNC:10001 NaN \n", "16424 HGNC:10001 NaN \n", "47487 HGNC:10001 NaN \n", "47488 HGNC:10001 NaN \n", "2684 HGNC:10004 biolink:gene_associated_with_condition \n", "... ... ... \n", "7192 NCBIGene:111365204 NaN \n", "2108 NCBIGene:111365204 biolink:gene_associated_with_condition \n", "9063 NCBIGene:111365204 NaN \n", "817 NCBIGene:7467 biolink:gene_associated_with_condition \n", "11624 NCBIGene:7467 NaN \n", "\n", " object primary_knowledge_source \\\n", "16423 MONDO:0007781 ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_... \n", "16424 MONDO:0007781 GENCC:000106 \n", "47487 MONDO:0007781 ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_... \n", "47488 MONDO:0007781 GENCC:000106 \n", "2684 MONDO:0012033 infores:medgen_mim_g2d \n", "... ... ... \n", "7192 MONDO:0007630 ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_... \n", "2108 MONDO:0010932 infores:medgen_mim_g2d \n", "9063 MONDO:0010932 ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_... \n", "817 MONDO:0008684 infores:medgen_mim_g2d \n", "11624 MONDO:0008684 ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_... \n", "\n", " aggregator_knowledge_source original_object original_subject \\\n", "16423 infores:hpoa OMIM:145500 NCBIGene:8490 \n", "16424 infores:gencc NaN NaN \n", "47487 infores:hpoa OMIM:145500 NCBIGene:8490 \n", "47488 infores:gencc NaN NaN \n", "2684 infores:medgen_mim_g2d OMIM:608415 NCBIGene:8787 \n", "... ... ... ... \n", "7192 infores:hpoa OMIM:136550 NaN \n", "2108 infores:medgen_mim_g2d OMIM:600790 NaN \n", "9063 infores:hpoa OMIM:600790 NaN \n", "817 infores:medgen_mim_g2d OMIM:194190 NaN \n", "11624 infores:hpoa OMIM:194190 NaN \n", "\n", " subject_prefix object_prefix subject_label \\\n", "16423 HGNC MONDO RGS5 \n", "16424 HGNC MONDO RGS5 \n", "47487 HGNC MONDO RGS5 \n", "47488 HGNC MONDO RGS5 \n", "2684 HGNC MONDO NaN \n", "... ... ... ... \n", "7192 NCBIGene MONDO - \n", "2108 NCBIGene MONDO NaN \n", "9063 NCBIGene MONDO - \n", "817 NCBIGene MONDO NaN \n", "11624 NCBIGene MONDO - \n", "\n", " object_label \n", "16423 NaN \n", "16424 essential hypertension, genetic \n", "47487 NaN \n", "47488 essential hypertension, genetic \n", "2684 NaN \n", "... ... \n", "7192 NaN \n", "2108 NaN \n", "9063 NaN \n", "817 NaN \n", "11624 NaN \n", "\n", "[106577 rows x 11 columns]" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.sort_values([\"subject\", \"object\"])" ] }, { "cell_type": "code", "execution_count": 16, "id": "f9ddd648", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
subjectobject
primary_knowledge_source
GENCC:00010199749974
GENCC:00010256875687
GENCC:00010449524952
GENCC:00010513651365
GENCC:00010691019101
GENCC:00010711451145
GENCC:000108767767
GENCC:0001101640716407
GENCC:00011136243624
GENCC:00011269186918
GENCC:0001137272
GENCC:0001144545
ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_medgen2036820368
http://www.orphadata.org/data/xml/en_product6.xml2024520245
infores:medgen_mim_g2d59075907
\n", "
" ], "text/plain": [ " subject object\n", "primary_knowledge_source \n", "GENCC:000101 9974 9974\n", "GENCC:000102 5687 5687\n", "GENCC:000104 4952 4952\n", "GENCC:000105 1365 1365\n", "GENCC:000106 9101 9101\n", "GENCC:000107 1145 1145\n", "GENCC:000108 767 767\n", "GENCC:000110 16407 16407\n", "GENCC:000111 3624 3624\n", "GENCC:000112 6918 6918\n", "GENCC:000113 72 72\n", "GENCC:000114 45 45\n", "ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/mim2gene_m... 20368 20368\n", "http://www.orphadata.org/data/xml/en_product6.xml 20245 20245\n", "infores:medgen_mim_g2d 5907 5907" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.groupby(\"primary_knowledge_source\")[[\"subject\", \"object\"]].count()" ] }, { "cell_type": "code", "execution_count": 25, "id": "a85d29d8-4106-45f7-b3b6-ad93a7f56262", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import matplotlib.pyplot as plt\n", "from matplotlib_venn import venn2, venn3\n", "\n", "s1 = \"infores:gencc\"\n", "s2 = \"infores:medgen_mim_g2d\"\n", "s3 = \"infores:hpoa\"\n", "src_col = \"aggregator_knowledge_source\"\n", "\n", "def make_venn(cols, venn_df=None):\n", " if venn_df is None:\n", " venn_df = df\n", " source1_tuples = set(venn_df[venn_df[src_col] == s1][cols].apply(tuple, axis=1))\n", " source2_tuples = set(venn_df[venn_df[src_col] == s2][cols].apply(tuple, axis=1))\n", " source3_tuples = set(venn_df[venn_df[src_col] == s3][cols].apply(tuple, axis=1))\n", " venn3([source1_tuples, source2_tuples, source3_tuples], set_labels=(s1, s2, s3))\n", " plt.title(f\"Venn Diagram for {cols}\")\n", " plt.show()\n", "\n", "# Identifying unique (a, b) tuples for each source\n", "#source1_tuples = set(df[df[src_col] == s1][['subject', 'object']].apply(tuple, axis=1))\n", "#source2_tuples = set(df[df[src_col] == s2][['subject', 'object']].apply(tuple, axis=1))\n", "\n", "# Creating the Venn diagram\n", "#venn2([source1_tuples, source2_tuples], set_labels=(s1, s2))\n", "\n", "#plt.title(\"Venn Diagram for gene-disease tuples\")\n", "#plt.show()\n", "\n", "make_venn([\"subject\", \"object\"])" ] }, { "cell_type": "code", "execution_count": 26, "id": "377bae54-27a4-4bb8-96ae-2e30c22fb87f", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "make_venn([\"subject\"])" ] }, { "cell_type": "code", "execution_count": 27, "id": "ba02dafd-08d0-4c4f-b9e9-83a7c769d277", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "make_venn([\"object\"])" ] }, { "cell_type": "code", "execution_count": 28, "id": "e9473a9d-a489-483c-b30e-99b8ed82c295", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "8697" ] }, "execution_count": 28, "metadata": {}, "output_type": "execute_result" } ], "source": [ "diseases = set(df[\"object\"])\n", "len(diseases)" ] }, { "cell_type": "code", "execution_count": 29, "id": "4c711ff2-1497-4dbb-9337-db7afc533475", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "1092" ] }, "execution_count": 29, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from oaklib.datamodels.vocabulary import IS_A\n", "\n", "redundant = set()\n", "roll_up_map = {}\n", "for d in diseases:\n", " ancs = adapter.ancestors(d, predicates=[IS_A], reflexive=False)\n", " ixn = diseases.intersection(ancs)\n", " redundant.update(ixn)\n", " if ixn:\n", " roll_up_map[d] = list(ixn)\n", " else:\n", " roll_up_map[d] = [d]\n", "\n", "len(redundant)" ] }, { "cell_type": "code", "execution_count": 30, "id": "548714e4-a057-465f-aed0-9e02baa36b5f", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0 (MONDO:0005240,)\n", "1 (MONDO:0005172, MONDO:0007034)\n", "2 (MONDO:0005172, MONDO:0005516)\n", "3 (MONDO:0005071, MONDO:0021061)\n", "4 (MONDO:0005172, MONDO:0015469)\n", " ... \n", "106572 (MONDO:0013673,)\n", "106573 (MONDO:0013673,)\n", "106574 (MONDO:0013673,)\n", "106575 (MONDO:0005046, MONDO:0021094)\n", "106576 (MONDO:0005046, MONDO:0021094)\n", "Name: object_rollup, Length: 106577, dtype: object" ] }, "execution_count": 30, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df['object_rollup'] = df['object'].map(lambda x: tuple(sorted(roll_up_map[x])))\n", "df[\"object_rollup\"]" ] }, { "cell_type": "code", "execution_count": 31, "id": "5298e865-6994-47e6-bcbb-7508aa3a8755", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "make_venn([\"subject\", \"object_rollup\"])" ] }, { "cell_type": "code", "execution_count": 15, "id": "c5f8a181", "metadata": {}, "outputs": [], "source": [ "cross_tab = pd.crosstab(df[df['primary_knowledge_source']==\"infores:gencc\"]['object'], df[df['primary_knowledge_source']==\"infores:medgen_mim_g2d\"]['object'])" ] }, { "cell_type": "code", "execution_count": 16, "id": "8e14b79b", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
object
object
\n", "
" ], "text/plain": [ "Empty DataFrame\n", "Columns: []\n", "Index: []" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "cross_tab" ] }, { "cell_type": "code", "execution_count": 11, "id": "46bc9cb1", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
subjectpredicateobjectprimary_knowledge_source
5837HGNC:10896NaNMONDO:0008426infores:gencc
5838HGNC:16636NaNMONDO:0008233infores:gencc
5839HGNC:16636NaNMONDO:0007308infores:gencc
5840HGNC:17939NaNMONDO:0044322infores:gencc
5841HGNC:11071NaNMONDO:0014572infores:gencc
...............
16089HGNC:17625NaNMONDO:0100284infores:gencc
16090HGNC:13128NaNMONDO:0100148infores:gencc
16091HGNC:13156NaNMONDO:0100284infores:gencc
16092HGNC:29046NaNMONDO:0030695infores:gencc
16093HGNC:6827NaNMONDO:0030770infores:gencc
\n", "

10257 rows × 4 columns

\n", "
" ], "text/plain": [ " subject predicate object primary_knowledge_source\n", "5837 HGNC:10896 NaN MONDO:0008426 infores:gencc\n", "5838 HGNC:16636 NaN MONDO:0008233 infores:gencc\n", "5839 HGNC:16636 NaN MONDO:0007308 infores:gencc\n", "5840 HGNC:17939 NaN MONDO:0044322 infores:gencc\n", "5841 HGNC:11071 NaN MONDO:0014572 infores:gencc\n", "... ... ... ... ...\n", "16089 HGNC:17625 NaN MONDO:0100284 infores:gencc\n", "16090 HGNC:13128 NaN MONDO:0100148 infores:gencc\n", "16091 HGNC:13156 NaN MONDO:0100284 infores:gencc\n", "16092 HGNC:29046 NaN MONDO:0030695 infores:gencc\n", "16093 HGNC:6827 NaN MONDO:0030770 infores:gencc\n", "\n", "[10257 rows x 4 columns]" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "gencc = df[df['primary_knowledge_source']==\"infores:gencc\"]\n", "gencc" ] }, { "cell_type": "code", "execution_count": 12, "id": "60d6657a", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
subjectpredicateobjectprimary_knowledge_source
0HGNC:1952biolink:gene_associated_with_conditionMONDO:0007032infores:medgen_mim_g2d
1HGNC:29216biolink:gene_associated_with_conditionMONDO:0024506infores:medgen_mim_g2d
2HGNC:3690biolink:gene_associated_with_conditionMONDO:0007037infores:medgen_mim_g2d
3HGNC:7773biolink:gene_associated_with_conditionMONDO:0007039infores:medgen_mim_g2d
4HGNC:3689biolink:gene_associated_with_conditionMONDO:0007041infores:medgen_mim_g2d
...............
5832HGNC:11016biolink:gene_associated_with_conditionOMIM:620306infores:medgen_mim_g2d
5833HGNC:10485biolink:gene_associated_with_conditionOMIM:620310infores:medgen_mim_g2d
5834HGNC:15979biolink:gene_associated_with_conditionOMIM:620311infores:medgen_mim_g2d
5835HGNC:19946biolink:gene_associated_with_conditionOMIM:620316infores:medgen_mim_g2d
5836HGNC:21839biolink:gene_associated_with_conditionOMIM:620319infores:medgen_mim_g2d
\n", "

5837 rows × 4 columns

\n", "
" ], "text/plain": [ " subject predicate object \\\n", "0 HGNC:1952 biolink:gene_associated_with_condition MONDO:0007032 \n", "1 HGNC:29216 biolink:gene_associated_with_condition MONDO:0024506 \n", "2 HGNC:3690 biolink:gene_associated_with_condition MONDO:0007037 \n", "3 HGNC:7773 biolink:gene_associated_with_condition MONDO:0007039 \n", "4 HGNC:3689 biolink:gene_associated_with_condition MONDO:0007041 \n", "... ... ... ... \n", "5832 HGNC:11016 biolink:gene_associated_with_condition OMIM:620306 \n", "5833 HGNC:10485 biolink:gene_associated_with_condition OMIM:620310 \n", "5834 HGNC:15979 biolink:gene_associated_with_condition OMIM:620311 \n", "5835 HGNC:19946 biolink:gene_associated_with_condition OMIM:620316 \n", "5836 HGNC:21839 biolink:gene_associated_with_condition OMIM:620319 \n", "\n", " primary_knowledge_source \n", "0 infores:medgen_mim_g2d \n", "1 infores:medgen_mim_g2d \n", "2 infores:medgen_mim_g2d \n", "3 infores:medgen_mim_g2d \n", "4 infores:medgen_mim_g2d \n", "... ... \n", "5832 infores:medgen_mim_g2d \n", "5833 infores:medgen_mim_g2d \n", "5834 infores:medgen_mim_g2d \n", "5835 infores:medgen_mim_g2d \n", "5836 infores:medgen_mim_g2d \n", "\n", "[5837 rows x 4 columns]" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "medgen = df[df['primary_knowledge_source']==\"infores:medgen_mim_g2d\"]\n", "medgen" ] }, { "cell_type": "code", "execution_count": 13, "id": "091b2eb2", "metadata": {}, "outputs": [ { "ename": "TypeError", "evalue": "crosstab() missing 1 required positional argument: 'columns'", "output_type": "error", "traceback": [ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mTypeError\u001b[0m Traceback (most recent call last)", "Cell \u001b[0;32mIn[13], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[43mpd\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcrosstab\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdf\u001b[49m\u001b[43m)\u001b[49m\n", "\u001b[0;31mTypeError\u001b[0m: crosstab() missing 1 required positional argument: 'columns'" ] } ], "source": [ "pd.crosstab(df)" ] }, { "cell_type": "code", "execution_count": null, "id": "74623173", "metadata": {}, "outputs": [], "source": [ "!pip install matplotlib-venn" ] }, { "cell_type": "code", "execution_count": null, "id": "7b4b173b", "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.9.5" } }, "nbformat": 4, "nbformat_minor": 5 }