[ { "name": "metadata-show-tables", "source": "metadata", "sql": "SHOW TABLES" }, { "name": "metadata-describe-cancer_study", "source": "metadata", "sql": "DESCRIBE TABLE cancer_study" }, { "name": "metadata-describe-genomic_event_derived", "source": "metadata", "sql": "DESCRIBE TABLE genomic_event_derived" }, { "name": "metadata-describe-sample_to_gene_panel_derived", "source": "metadata", "sql": "DESCRIBE TABLE sample_to_gene_panel_derived" }, { "name": "metadata-describe-clinical_data_derived", "source": "metadata", "sql": "DESCRIBE TABLE clinical_data_derived" }, { "name": "metadata-describe-sample_derived", "source": "metadata", "sql": "DESCRIBE TABLE sample_derived" }, { "name": "metadata-describe-gene", "source": "metadata", "sql": "DESCRIBE TABLE gene" }, { "name": "metadata-describe-genetic_profile", "source": "metadata", "sql": "DESCRIBE TABLE genetic_profile" }, { "name": "metadata-version-timezone", "source": "metadata", "sql": "SELECT version(), timezone()" }, { "name": "metadata-settings", "source": "metadata", "sql": "SELECT name, value FROM system.settings LIMIT 1000" }, { "name": "inventory", "source": "metadata", "sql": "SELECT name, total_rows, total_bytes FROM system.tables WHERE database=currentDatabase() ORDER BY name" }, { "name": "lookup-estimate-gene", "source": "metadata", "sql": "SELECT hugo_gene_symbol, entrez_gene_id FROM gene WHERE hugo_gene_symbol='TP53'" }, { "name": "lookup-estimate-study", "source": "metadata", "sql": "SELECT cancer_study_identifier FROM cancer_study WHERE cancer_study_identifier='msk_chord_2024'" }, { "name": "160-before-TP53-mutation", "source": "pr160", "sql": "SELECT * FROM gene_alteration_frequency_by_cancer_type(preference='all_studies_non_redundant', gene='TP53', alteration='mutation')" }, { "name": "160-after-TP53-mutation", "source": "pr160", "sql": "WITH cohort AS (\n SELECT cancer_study_identifier\n FROM cancer_study_query_preferences\n WHERE preference_name = 'all_studies_non_redundant'\n),\nsample_cancer_type AS (\n SELECT cd.sample_unique_id, cd.attribute_value AS cancer_type\n FROM clinical_data_derived cd\n JOIN cohort c USING (cancer_study_identifier)\n WHERE cd.attribute_name = 'CANCER_TYPE'\n),\naltered AS (\n SELECT sct.cancer_type,\n COUNT(DISTINCT ged.sample_unique_id) AS altered_samples\n FROM genomic_event_derived ged\n JOIN cohort c USING (cancer_study_identifier)\n JOIN sample_cancer_type sct USING (sample_unique_id)\n WHERE ged.hugo_gene_symbol = 'TP53'\n AND ged.off_panel = 0\n AND (\n ('mutation' = 'mutation'\n AND ged.variant_type = 'mutation'\n AND ged.mutation_status != 'UNCALLED')\n OR ('mutation' = 'amplification'\n AND ged.variant_type = 'cna'\n AND ged.cna_alteration = 2)\n OR ('mutation' = 'deep_deletion'\n AND ged.variant_type = 'cna'\n AND ged.cna_alteration = -2)\n OR ('mutation' = 'structural_variant'\n AND ged.variant_type = 'structural_variant')\n )\n GROUP BY sct.cancer_type\n),\nprofiled_samples_for_gene AS (\n -- Map the user-facing alteration token to the alteration_type stored\n -- on sample_to_gene_panel_derived. Same gene-in-panel-or-WES branch\n -- as the mutation view, but with the matching alteration_type filter.\n SELECT stgp.sample_unique_id, stgp.cancer_study_identifier\n FROM sample_to_gene_panel_derived stgp\n JOIN gene_panel gp ON stgp.gene_panel_id = gp.stable_id\n JOIN gene_panel_list gpl ON gp.internal_id = gpl.internal_id\n -- IN-subquery instead of JOIN gene: resolves the symbol to its entrez\n -- id(s) once, rather than joining every panel row against the gene\n -- table before filtering. COUNT(DISTINCT) downstream makes the result\n -- identical even for symbols with several (or duplicated) gene rows.\n -- IS NOT NULL keeps NULL keys unmatched (as the equality JOIN did) even\n -- under transform_null_in=1.\n WHERE gpl.gene_id IN (\n SELECT entrez_gene_id FROM gene\n WHERE hugo_gene_symbol = 'TP53' AND entrez_gene_id IS NOT NULL)\n AND stgp.alteration_type = multiIf(\n 'mutation' = 'mutation', 'MUTATION_EXTENDED',\n 'mutation' = 'amplification', 'COPY_NUMBER_ALTERATION',\n 'mutation' = 'deep_deletion', 'COPY_NUMBER_ALTERATION',\n 'mutation' = 'structural_variant', 'STRUCTURAL_VARIANT',\n '')\n UNION ALL\n SELECT sample_unique_id, cancer_study_identifier\n FROM sample_to_gene_panel_derived\n WHERE gene_panel_id = 'WES'\n AND alteration_type = multiIf(\n 'mutation' = 'mutation', 'MUTATION_EXTENDED',\n 'mutation' = 'amplification', 'COPY_NUMBER_ALTERATION',\n 'mutation' = 'deep_deletion', 'COPY_NUMBER_ALTERATION',\n 'mutation' = 'structural_variant', 'STRUCTURAL_VARIANT',\n '')\n),\nprofiled AS (\n SELECT sct.cancer_type,\n COUNT(DISTINCT p.sample_unique_id) AS profiled_samples\n FROM profiled_samples_for_gene p\n JOIN cohort c USING (cancer_study_identifier)\n JOIN sample_cancer_type sct USING (sample_unique_id)\n GROUP BY sct.cancer_type\n)\nSELECT a.cancer_type,\n a.altered_samples,\n p.profiled_samples,\n ROUND(a.altered_samples * 100.0 / NULLIF(p.profiled_samples, 0), 1) AS frequency_pct\nFROM altered a\nJOIN profiled p USING (cancer_type)\nWHERE p.profiled_samples >= 50" }, { "name": "160-before-TP53-amplification", "source": "pr160", "sql": "SELECT * FROM gene_alteration_frequency_by_cancer_type(preference='all_studies_non_redundant', gene='TP53', alteration='amplification')" }, { "name": "160-after-TP53-amplification", "source": "pr160", "sql": "WITH cohort AS (\n SELECT cancer_study_identifier\n FROM cancer_study_query_preferences\n WHERE preference_name = 'all_studies_non_redundant'\n),\nsample_cancer_type AS (\n SELECT cd.sample_unique_id, cd.attribute_value AS cancer_type\n FROM clinical_data_derived cd\n JOIN cohort c USING (cancer_study_identifier)\n WHERE cd.attribute_name = 'CANCER_TYPE'\n),\naltered AS (\n SELECT sct.cancer_type,\n COUNT(DISTINCT ged.sample_unique_id) AS altered_samples\n FROM genomic_event_derived ged\n JOIN cohort c USING (cancer_study_identifier)\n JOIN sample_cancer_type sct USING (sample_unique_id)\n WHERE ged.hugo_gene_symbol = 'TP53'\n AND ged.off_panel = 0\n AND (\n ('amplification' = 'mutation'\n AND ged.variant_type = 'mutation'\n AND ged.mutation_status != 'UNCALLED')\n OR ('amplification' = 'amplification'\n AND ged.variant_type = 'cna'\n AND ged.cna_alteration = 2)\n OR ('amplification' = 'deep_deletion'\n AND ged.variant_type = 'cna'\n AND ged.cna_alteration = -2)\n OR ('amplification' = 'structural_variant'\n AND ged.variant_type = 'structural_variant')\n )\n GROUP BY sct.cancer_type\n),\nprofiled_samples_for_gene AS (\n -- Map the user-facing alteration token to the alteration_type stored\n -- on sample_to_gene_panel_derived. Same gene-in-panel-or-WES branch\n -- as the mutation view, but with the matching alteration_type filter.\n SELECT stgp.sample_unique_id, stgp.cancer_study_identifier\n FROM sample_to_gene_panel_derived stgp\n JOIN gene_panel gp ON stgp.gene_panel_id = gp.stable_id\n JOIN gene_panel_list gpl ON gp.internal_id = gpl.internal_id\n -- IN-subquery instead of JOIN gene: resolves the symbol to its entrez\n -- id(s) once, rather than joining every panel row against the gene\n -- table before filtering. COUNT(DISTINCT) downstream makes the result\n -- identical even for symbols with several (or duplicated) gene rows.\n -- IS NOT NULL keeps NULL keys unmatched (as the equality JOIN did) even\n -- under transform_null_in=1.\n WHERE gpl.gene_id IN (\n SELECT entrez_gene_id FROM gene\n WHERE hugo_gene_symbol = 'TP53' AND entrez_gene_id IS NOT NULL)\n AND stgp.alteration_type = multiIf(\n 'amplification' = 'mutation', 'MUTATION_EXTENDED',\n 'amplification' = 'amplification', 'COPY_NUMBER_ALTERATION',\n 'amplification' = 'deep_deletion', 'COPY_NUMBER_ALTERATION',\n 'amplification' = 'structural_variant', 'STRUCTURAL_VARIANT',\n '')\n UNION ALL\n SELECT sample_unique_id, cancer_study_identifier\n FROM sample_to_gene_panel_derived\n WHERE gene_panel_id = 'WES'\n AND alteration_type = multiIf(\n 'amplification' = 'mutation', 'MUTATION_EXTENDED',\n 'amplification' = 'amplification', 'COPY_NUMBER_ALTERATION',\n 'amplification' = 'deep_deletion', 'COPY_NUMBER_ALTERATION',\n 'amplification' = 'structural_variant', 'STRUCTURAL_VARIANT',\n '')\n),\nprofiled AS (\n SELECT sct.cancer_type,\n COUNT(DISTINCT p.sample_unique_id) AS profiled_samples\n FROM profiled_samples_for_gene p\n JOIN cohort c USING (cancer_study_identifier)\n JOIN sample_cancer_type sct USING (sample_unique_id)\n GROUP BY sct.cancer_type\n)\nSELECT a.cancer_type,\n a.altered_samples,\n p.profiled_samples,\n ROUND(a.altered_samples * 100.0 / NULLIF(p.profiled_samples, 0), 1) AS frequency_pct\nFROM altered a\nJOIN profiled p USING (cancer_type)\nWHERE p.profiled_samples >= 50" }, { "name": "160-before-TP53-deep_deletion", "source": "pr160", "sql": "SELECT * FROM gene_alteration_frequency_by_cancer_type(preference='all_studies_non_redundant', gene='TP53', alteration='deep_deletion')" }, { "name": "160-after-TP53-deep_deletion", "source": "pr160", "sql": "WITH cohort AS (\n SELECT cancer_study_identifier\n FROM cancer_study_query_preferences\n WHERE preference_name = 'all_studies_non_redundant'\n),\nsample_cancer_type AS (\n SELECT cd.sample_unique_id, cd.attribute_value AS cancer_type\n FROM clinical_data_derived cd\n JOIN cohort c USING (cancer_study_identifier)\n WHERE cd.attribute_name = 'CANCER_TYPE'\n),\naltered AS (\n SELECT sct.cancer_type,\n COUNT(DISTINCT ged.sample_unique_id) AS altered_samples\n FROM genomic_event_derived ged\n JOIN cohort c USING (cancer_study_identifier)\n JOIN sample_cancer_type sct USING (sample_unique_id)\n WHERE ged.hugo_gene_symbol = 'TP53'\n AND ged.off_panel = 0\n AND (\n ('deep_deletion' = 'mutation'\n AND ged.variant_type = 'mutation'\n AND ged.mutation_status != 'UNCALLED')\n OR ('deep_deletion' = 'amplification'\n AND ged.variant_type = 'cna'\n AND ged.cna_alteration = 2)\n OR ('deep_deletion' = 'deep_deletion'\n AND ged.variant_type = 'cna'\n AND ged.cna_alteration = -2)\n OR ('deep_deletion' = 'structural_variant'\n AND ged.variant_type = 'structural_variant')\n )\n GROUP BY sct.cancer_type\n),\nprofiled_samples_for_gene AS (\n -- Map the user-facing alteration token to the alteration_type stored\n -- on sample_to_gene_panel_derived. Same gene-in-panel-or-WES branch\n -- as the mutation view, but with the matching alteration_type filter.\n SELECT stgp.sample_unique_id, stgp.cancer_study_identifier\n FROM sample_to_gene_panel_derived stgp\n JOIN gene_panel gp ON stgp.gene_panel_id = gp.stable_id\n JOIN gene_panel_list gpl ON gp.internal_id = gpl.internal_id\n -- IN-subquery instead of JOIN gene: resolves the symbol to its entrez\n -- id(s) once, rather than joining every panel row against the gene\n -- table before filtering. COUNT(DISTINCT) downstream makes the result\n -- identical even for symbols with several (or duplicated) gene rows.\n -- IS NOT NULL keeps NULL keys unmatched (as the equality JOIN did) even\n -- under transform_null_in=1.\n WHERE gpl.gene_id IN (\n SELECT entrez_gene_id FROM gene\n WHERE hugo_gene_symbol = 'TP53' AND entrez_gene_id IS NOT NULL)\n AND stgp.alteration_type = multiIf(\n 'deep_deletion' = 'mutation', 'MUTATION_EXTENDED',\n 'deep_deletion' = 'amplification', 'COPY_NUMBER_ALTERATION',\n 'deep_deletion' = 'deep_deletion', 'COPY_NUMBER_ALTERATION',\n 'deep_deletion' = 'structural_variant', 'STRUCTURAL_VARIANT',\n '')\n UNION ALL\n SELECT sample_unique_id, cancer_study_identifier\n FROM sample_to_gene_panel_derived\n WHERE gene_panel_id = 'WES'\n AND alteration_type = multiIf(\n 'deep_deletion' = 'mutation', 'MUTATION_EXTENDED',\n 'deep_deletion' = 'amplification', 'COPY_NUMBER_ALTERATION',\n 'deep_deletion' = 'deep_deletion', 'COPY_NUMBER_ALTERATION',\n 'deep_deletion' = 'structural_variant', 'STRUCTURAL_VARIANT',\n '')\n),\nprofiled AS (\n SELECT sct.cancer_type,\n COUNT(DISTINCT p.sample_unique_id) AS profiled_samples\n FROM profiled_samples_for_gene p\n JOIN cohort c USING (cancer_study_identifier)\n JOIN sample_cancer_type sct USING (sample_unique_id)\n GROUP BY sct.cancer_type\n)\nSELECT a.cancer_type,\n a.altered_samples,\n p.profiled_samples,\n ROUND(a.altered_samples * 100.0 / NULLIF(p.profiled_samples, 0), 1) AS frequency_pct\nFROM altered a\nJOIN profiled p USING (cancer_type)\nWHERE p.profiled_samples >= 50" }, { "name": "160-before-TP53-structural_variant", "source": "pr160", "sql": "SELECT * FROM gene_alteration_frequency_by_cancer_type(preference='all_studies_non_redundant', gene='TP53', alteration='structural_variant')" }, { "name": "160-after-TP53-structural_variant", "source": "pr160", "sql": "WITH cohort AS (\n SELECT cancer_study_identifier\n FROM cancer_study_query_preferences\n WHERE preference_name = 'all_studies_non_redundant'\n),\nsample_cancer_type AS (\n SELECT cd.sample_unique_id, cd.attribute_value AS cancer_type\n FROM clinical_data_derived cd\n JOIN cohort c USING (cancer_study_identifier)\n WHERE cd.attribute_name = 'CANCER_TYPE'\n),\naltered AS (\n SELECT sct.cancer_type,\n COUNT(DISTINCT ged.sample_unique_id) AS altered_samples\n FROM genomic_event_derived ged\n JOIN cohort c USING (cancer_study_identifier)\n JOIN sample_cancer_type sct USING (sample_unique_id)\n WHERE ged.hugo_gene_symbol = 'TP53'\n AND ged.off_panel = 0\n AND (\n ('structural_variant' = 'mutation'\n AND ged.variant_type = 'mutation'\n AND ged.mutation_status != 'UNCALLED')\n OR ('structural_variant' = 'amplification'\n AND ged.variant_type = 'cna'\n AND ged.cna_alteration = 2)\n OR ('structural_variant' = 'deep_deletion'\n AND ged.variant_type = 'cna'\n AND ged.cna_alteration = -2)\n OR ('structural_variant' = 'structural_variant'\n AND ged.variant_type = 'structural_variant')\n )\n GROUP BY sct.cancer_type\n),\nprofiled_samples_for_gene AS (\n -- Map the user-facing alteration token to the alteration_type stored\n -- on sample_to_gene_panel_derived. Same gene-in-panel-or-WES branch\n -- as the mutation view, but with the matching alteration_type filter.\n SELECT stgp.sample_unique_id, stgp.cancer_study_identifier\n FROM sample_to_gene_panel_derived stgp\n JOIN gene_panel gp ON stgp.gene_panel_id = gp.stable_id\n JOIN gene_panel_list gpl ON gp.internal_id = gpl.internal_id\n -- IN-subquery instead of JOIN gene: resolves the symbol to its entrez\n -- id(s) once, rather than joining every panel row against the gene\n -- table before filtering. COUNT(DISTINCT) downstream makes the result\n -- identical even for symbols with several (or duplicated) gene rows.\n -- IS NOT NULL keeps NULL keys unmatched (as the equality JOIN did) even\n -- under transform_null_in=1.\n WHERE gpl.gene_id IN (\n SELECT entrez_gene_id FROM gene\n WHERE hugo_gene_symbol = 'TP53' AND entrez_gene_id IS NOT NULL)\n AND stgp.alteration_type = multiIf(\n 'structural_variant' = 'mutation', 'MUTATION_EXTENDED',\n 'structural_variant' = 'amplification', 'COPY_NUMBER_ALTERATION',\n 'structural_variant' = 'deep_deletion', 'COPY_NUMBER_ALTERATION',\n 'structural_variant' = 'structural_variant', 'STRUCTURAL_VARIANT',\n '')\n UNION ALL\n SELECT sample_unique_id, cancer_study_identifier\n FROM sample_to_gene_panel_derived\n WHERE gene_panel_id = 'WES'\n AND alteration_type = multiIf(\n 'structural_variant' = 'mutation', 'MUTATION_EXTENDED',\n 'structural_variant' = 'amplification', 'COPY_NUMBER_ALTERATION',\n 'structural_variant' = 'deep_deletion', 'COPY_NUMBER_ALTERATION',\n 'structural_variant' = 'structural_variant', 'STRUCTURAL_VARIANT',\n '')\n),\nprofiled AS (\n SELECT sct.cancer_type,\n COUNT(DISTINCT p.sample_unique_id) AS profiled_samples\n FROM profiled_samples_for_gene p\n JOIN cohort c USING (cancer_study_identifier)\n JOIN sample_cancer_type sct USING (sample_unique_id)\n GROUP BY sct.cancer_type\n)\nSELECT a.cancer_type,\n a.altered_samples,\n p.profiled_samples,\n ROUND(a.altered_samples * 100.0 / NULLIF(p.profiled_samples, 0), 1) AS frequency_pct\nFROM altered a\nJOIN profiled p USING (cancer_type)\nWHERE p.profiled_samples >= 50" }, { "name": "160-before-ERBB2-mutation", "source": "pr160", "sql": "SELECT * FROM gene_alteration_frequency_by_cancer_type(preference='all_studies_non_redundant', gene='ERBB2', alteration='mutation')" }, { "name": "160-after-ERBB2-mutation", "source": "pr160", "sql": "WITH cohort AS (\n SELECT cancer_study_identifier\n FROM cancer_study_query_preferences\n WHERE preference_name = 'all_studies_non_redundant'\n),\nsample_cancer_type AS (\n SELECT cd.sample_unique_id, cd.attribute_value AS cancer_type\n FROM clinical_data_derived cd\n JOIN cohort c USING (cancer_study_identifier)\n WHERE cd.attribute_name = 'CANCER_TYPE'\n),\naltered AS (\n SELECT sct.cancer_type,\n COUNT(DISTINCT ged.sample_unique_id) AS altered_samples\n FROM genomic_event_derived ged\n JOIN cohort c USING (cancer_study_identifier)\n JOIN sample_cancer_type sct USING (sample_unique_id)\n WHERE ged.hugo_gene_symbol = 'ERBB2'\n AND ged.off_panel = 0\n AND (\n ('mutation' = 'mutation'\n AND ged.variant_type = 'mutation'\n AND ged.mutation_status != 'UNCALLED')\n OR ('mutation' = 'amplification'\n AND ged.variant_type = 'cna'\n AND ged.cna_alteration = 2)\n OR ('mutation' = 'deep_deletion'\n AND ged.variant_type = 'cna'\n AND ged.cna_alteration = -2)\n OR ('mutation' = 'structural_variant'\n AND ged.variant_type = 'structural_variant')\n )\n GROUP BY sct.cancer_type\n),\nprofiled_samples_for_gene AS (\n -- Map the user-facing alteration token to the alteration_type stored\n -- on sample_to_gene_panel_derived. Same gene-in-panel-or-WES branch\n -- as the mutation view, but with the matching alteration_type filter.\n SELECT stgp.sample_unique_id, stgp.cancer_study_identifier\n FROM sample_to_gene_panel_derived stgp\n JOIN gene_panel gp ON stgp.gene_panel_id = gp.stable_id\n JOIN gene_panel_list gpl ON gp.internal_id = gpl.internal_id\n -- IN-subquery instead of JOIN gene: resolves the symbol to its entrez\n -- id(s) once, rather than joining every panel row against the gene\n -- table before filtering. COUNT(DISTINCT) downstream makes the result\n -- identical even for symbols with several (or duplicated) gene rows.\n -- IS NOT NULL keeps NULL keys unmatched (as the equality JOIN did) even\n -- under transform_null_in=1.\n WHERE gpl.gene_id IN (\n SELECT entrez_gene_id FROM gene\n WHERE hugo_gene_symbol = 'ERBB2' AND entrez_gene_id IS NOT NULL)\n AND stgp.alteration_type = multiIf(\n 'mutation' = 'mutation', 'MUTATION_EXTENDED',\n 'mutation' = 'amplification', 'COPY_NUMBER_ALTERATION',\n 'mutation' = 'deep_deletion', 'COPY_NUMBER_ALTERATION',\n 'mutation' = 'structural_variant', 'STRUCTURAL_VARIANT',\n '')\n UNION ALL\n SELECT sample_unique_id, cancer_study_identifier\n FROM sample_to_gene_panel_derived\n WHERE gene_panel_id = 'WES'\n AND alteration_type = multiIf(\n 'mutation' = 'mutation', 'MUTATION_EXTENDED',\n 'mutation' = 'amplification', 'COPY_NUMBER_ALTERATION',\n 'mutation' = 'deep_deletion', 'COPY_NUMBER_ALTERATION',\n 'mutation' = 'structural_variant', 'STRUCTURAL_VARIANT',\n '')\n),\nprofiled AS (\n SELECT sct.cancer_type,\n COUNT(DISTINCT p.sample_unique_id) AS profiled_samples\n FROM profiled_samples_for_gene p\n JOIN cohort c USING (cancer_study_identifier)\n JOIN sample_cancer_type sct USING (sample_unique_id)\n GROUP BY sct.cancer_type\n)\nSELECT a.cancer_type,\n a.altered_samples,\n p.profiled_samples,\n ROUND(a.altered_samples * 100.0 / NULLIF(p.profiled_samples, 0), 1) AS frequency_pct\nFROM altered a\nJOIN profiled p USING (cancer_type)\nWHERE p.profiled_samples >= 50" }, { "name": "160-before-ERBB2-amplification", "source": "pr160", "sql": "SELECT * FROM gene_alteration_frequency_by_cancer_type(preference='all_studies_non_redundant', gene='ERBB2', alteration='amplification')" }, { "name": "160-after-ERBB2-amplification", "source": "pr160", "sql": "WITH cohort AS (\n SELECT cancer_study_identifier\n FROM cancer_study_query_preferences\n WHERE preference_name = 'all_studies_non_redundant'\n),\nsample_cancer_type AS (\n SELECT cd.sample_unique_id, cd.attribute_value AS cancer_type\n FROM clinical_data_derived cd\n JOIN cohort c USING (cancer_study_identifier)\n WHERE cd.attribute_name = 'CANCER_TYPE'\n),\naltered AS (\n SELECT sct.cancer_type,\n COUNT(DISTINCT ged.sample_unique_id) AS altered_samples\n FROM genomic_event_derived ged\n JOIN cohort c USING (cancer_study_identifier)\n JOIN sample_cancer_type sct USING (sample_unique_id)\n WHERE ged.hugo_gene_symbol = 'ERBB2'\n AND ged.off_panel = 0\n AND (\n ('amplification' = 'mutation'\n AND ged.variant_type = 'mutation'\n AND ged.mutation_status != 'UNCALLED')\n OR ('amplification' = 'amplification'\n AND ged.variant_type = 'cna'\n AND ged.cna_alteration = 2)\n OR ('amplification' = 'deep_deletion'\n AND ged.variant_type = 'cna'\n AND ged.cna_alteration = -2)\n OR ('amplification' = 'structural_variant'\n AND ged.variant_type = 'structural_variant')\n )\n GROUP BY sct.cancer_type\n),\nprofiled_samples_for_gene AS (\n -- Map the user-facing alteration token to the alteration_type stored\n -- on sample_to_gene_panel_derived. Same gene-in-panel-or-WES branch\n -- as the mutation view, but with the matching alteration_type filter.\n SELECT stgp.sample_unique_id, stgp.cancer_study_identifier\n FROM sample_to_gene_panel_derived stgp\n JOIN gene_panel gp ON stgp.gene_panel_id = gp.stable_id\n JOIN gene_panel_list gpl ON gp.internal_id = gpl.internal_id\n -- IN-subquery instead of JOIN gene: resolves the symbol to its entrez\n -- id(s) once, rather than joining every panel row against the gene\n -- table before filtering. COUNT(DISTINCT) downstream makes the result\n -- identical even for symbols with several (or duplicated) gene rows.\n -- IS NOT NULL keeps NULL keys unmatched (as the equality JOIN did) even\n -- under transform_null_in=1.\n WHERE gpl.gene_id IN (\n SELECT entrez_gene_id FROM gene\n WHERE hugo_gene_symbol = 'ERBB2' AND entrez_gene_id IS NOT NULL)\n AND stgp.alteration_type = multiIf(\n 'amplification' = 'mutation', 'MUTATION_EXTENDED',\n 'amplification' = 'amplification', 'COPY_NUMBER_ALTERATION',\n 'amplification' = 'deep_deletion', 'COPY_NUMBER_ALTERATION',\n 'amplification' = 'structural_variant', 'STRUCTURAL_VARIANT',\n '')\n UNION ALL\n SELECT sample_unique_id, cancer_study_identifier\n FROM sample_to_gene_panel_derived\n WHERE gene_panel_id = 'WES'\n AND alteration_type = multiIf(\n 'amplification' = 'mutation', 'MUTATION_EXTENDED',\n 'amplification' = 'amplification', 'COPY_NUMBER_ALTERATION',\n 'amplification' = 'deep_deletion', 'COPY_NUMBER_ALTERATION',\n 'amplification' = 'structural_variant', 'STRUCTURAL_VARIANT',\n '')\n),\nprofiled AS (\n SELECT sct.cancer_type,\n COUNT(DISTINCT p.sample_unique_id) AS profiled_samples\n FROM profiled_samples_for_gene p\n JOIN cohort c USING (cancer_study_identifier)\n JOIN sample_cancer_type sct USING (sample_unique_id)\n GROUP BY sct.cancer_type\n)\nSELECT a.cancer_type,\n a.altered_samples,\n p.profiled_samples,\n ROUND(a.altered_samples * 100.0 / NULLIF(p.profiled_samples, 0), 1) AS frequency_pct\nFROM altered a\nJOIN profiled p USING (cancer_type)\nWHERE p.profiled_samples >= 50" }, { "name": "160-before-ERBB2-deep_deletion", "source": "pr160", "sql": "SELECT * FROM gene_alteration_frequency_by_cancer_type(preference='all_studies_non_redundant', gene='ERBB2', alteration='deep_deletion')" }, { "name": "160-after-ERBB2-deep_deletion", "source": "pr160", "sql": "WITH cohort AS (\n SELECT cancer_study_identifier\n FROM cancer_study_query_preferences\n WHERE preference_name = 'all_studies_non_redundant'\n),\nsample_cancer_type AS (\n SELECT cd.sample_unique_id, cd.attribute_value AS cancer_type\n FROM clinical_data_derived cd\n JOIN cohort c USING (cancer_study_identifier)\n WHERE cd.attribute_name = 'CANCER_TYPE'\n),\naltered AS (\n SELECT sct.cancer_type,\n COUNT(DISTINCT ged.sample_unique_id) AS altered_samples\n FROM genomic_event_derived ged\n JOIN cohort c USING (cancer_study_identifier)\n JOIN sample_cancer_type sct USING (sample_unique_id)\n WHERE ged.hugo_gene_symbol = 'ERBB2'\n AND ged.off_panel = 0\n AND (\n ('deep_deletion' = 'mutation'\n AND ged.variant_type = 'mutation'\n AND ged.mutation_status != 'UNCALLED')\n OR ('deep_deletion' = 'amplification'\n AND ged.variant_type = 'cna'\n AND ged.cna_alteration = 2)\n OR ('deep_deletion' = 'deep_deletion'\n AND ged.variant_type = 'cna'\n AND ged.cna_alteration = -2)\n OR ('deep_deletion' = 'structural_variant'\n AND ged.variant_type = 'structural_variant')\n )\n GROUP BY sct.cancer_type\n),\nprofiled_samples_for_gene AS (\n -- Map the user-facing alteration token to the alteration_type stored\n -- on sample_to_gene_panel_derived. Same gene-in-panel-or-WES branch\n -- as the mutation view, but with the matching alteration_type filter.\n SELECT stgp.sample_unique_id, stgp.cancer_study_identifier\n FROM sample_to_gene_panel_derived stgp\n JOIN gene_panel gp ON stgp.gene_panel_id = gp.stable_id\n JOIN gene_panel_list gpl ON gp.internal_id = gpl.internal_id\n -- IN-subquery instead of JOIN gene: resolves the symbol to its entrez\n -- id(s) once, rather than joining every panel row against the gene\n -- table before filtering. COUNT(DISTINCT) downstream makes the result\n -- identical even for symbols with several (or duplicated) gene rows.\n -- IS NOT NULL keeps NULL keys unmatched (as the equality JOIN did) even\n -- under transform_null_in=1.\n WHERE gpl.gene_id IN (\n SELECT entrez_gene_id FROM gene\n WHERE hugo_gene_symbol = 'ERBB2' AND entrez_gene_id IS NOT NULL)\n AND stgp.alteration_type = multiIf(\n 'deep_deletion' = 'mutation', 'MUTATION_EXTENDED',\n 'deep_deletion' = 'amplification', 'COPY_NUMBER_ALTERATION',\n 'deep_deletion' = 'deep_deletion', 'COPY_NUMBER_ALTERATION',\n 'deep_deletion' = 'structural_variant', 'STRUCTURAL_VARIANT',\n '')\n UNION ALL\n SELECT sample_unique_id, cancer_study_identifier\n FROM sample_to_gene_panel_derived\n WHERE gene_panel_id = 'WES'\n AND alteration_type = multiIf(\n 'deep_deletion' = 'mutation', 'MUTATION_EXTENDED',\n 'deep_deletion' = 'amplification', 'COPY_NUMBER_ALTERATION',\n 'deep_deletion' = 'deep_deletion', 'COPY_NUMBER_ALTERATION',\n 'deep_deletion' = 'structural_variant', 'STRUCTURAL_VARIANT',\n '')\n),\nprofiled AS (\n SELECT sct.cancer_type,\n COUNT(DISTINCT p.sample_unique_id) AS profiled_samples\n FROM profiled_samples_for_gene p\n JOIN cohort c USING (cancer_study_identifier)\n JOIN sample_cancer_type sct USING (sample_unique_id)\n GROUP BY sct.cancer_type\n)\nSELECT a.cancer_type,\n a.altered_samples,\n p.profiled_samples,\n ROUND(a.altered_samples * 100.0 / NULLIF(p.profiled_samples, 0), 1) AS frequency_pct\nFROM altered a\nJOIN profiled p USING (cancer_type)\nWHERE p.profiled_samples >= 50" }, { "name": "160-before-ERBB2-structural_variant", "source": "pr160", "sql": "SELECT * FROM gene_alteration_frequency_by_cancer_type(preference='all_studies_non_redundant', gene='ERBB2', alteration='structural_variant')" }, { "name": "160-after-ERBB2-structural_variant", "source": "pr160", "sql": "WITH cohort AS (\n SELECT cancer_study_identifier\n FROM cancer_study_query_preferences\n WHERE preference_name = 'all_studies_non_redundant'\n),\nsample_cancer_type AS (\n SELECT cd.sample_unique_id, cd.attribute_value AS cancer_type\n FROM clinical_data_derived cd\n JOIN cohort c USING (cancer_study_identifier)\n WHERE cd.attribute_name = 'CANCER_TYPE'\n),\naltered AS (\n SELECT sct.cancer_type,\n COUNT(DISTINCT ged.sample_unique_id) AS altered_samples\n FROM genomic_event_derived ged\n JOIN cohort c USING (cancer_study_identifier)\n JOIN sample_cancer_type sct USING (sample_unique_id)\n WHERE ged.hugo_gene_symbol = 'ERBB2'\n AND ged.off_panel = 0\n AND (\n ('structural_variant' = 'mutation'\n AND ged.variant_type = 'mutation'\n AND ged.mutation_status != 'UNCALLED')\n OR ('structural_variant' = 'amplification'\n AND ged.variant_type = 'cna'\n AND ged.cna_alteration = 2)\n OR ('structural_variant' = 'deep_deletion'\n AND ged.variant_type = 'cna'\n AND ged.cna_alteration = -2)\n OR ('structural_variant' = 'structural_variant'\n AND ged.variant_type = 'structural_variant')\n )\n GROUP BY sct.cancer_type\n),\nprofiled_samples_for_gene AS (\n -- Map the user-facing alteration token to the alteration_type stored\n -- on sample_to_gene_panel_derived. Same gene-in-panel-or-WES branch\n -- as the mutation view, but with the matching alteration_type filter.\n SELECT stgp.sample_unique_id, stgp.cancer_study_identifier\n FROM sample_to_gene_panel_derived stgp\n JOIN gene_panel gp ON stgp.gene_panel_id = gp.stable_id\n JOIN gene_panel_list gpl ON gp.internal_id = gpl.internal_id\n -- IN-subquery instead of JOIN gene: resolves the symbol to its entrez\n -- id(s) once, rather than joining every panel row against the gene\n -- table before filtering. COUNT(DISTINCT) downstream makes the result\n -- identical even for symbols with several (or duplicated) gene rows.\n -- IS NOT NULL keeps NULL keys unmatched (as the equality JOIN did) even\n -- under transform_null_in=1.\n WHERE gpl.gene_id IN (\n SELECT entrez_gene_id FROM gene\n WHERE hugo_gene_symbol = 'ERBB2' AND entrez_gene_id IS NOT NULL)\n AND stgp.alteration_type = multiIf(\n 'structural_variant' = 'mutation', 'MUTATION_EXTENDED',\n 'structural_variant' = 'amplification', 'COPY_NUMBER_ALTERATION',\n 'structural_variant' = 'deep_deletion', 'COPY_NUMBER_ALTERATION',\n 'structural_variant' = 'structural_variant', 'STRUCTURAL_VARIANT',\n '')\n UNION ALL\n SELECT sample_unique_id, cancer_study_identifier\n FROM sample_to_gene_panel_derived\n WHERE gene_panel_id = 'WES'\n AND alteration_type = multiIf(\n 'structural_variant' = 'mutation', 'MUTATION_EXTENDED',\n 'structural_variant' = 'amplification', 'COPY_NUMBER_ALTERATION',\n 'structural_variant' = 'deep_deletion', 'COPY_NUMBER_ALTERATION',\n 'structural_variant' = 'structural_variant', 'STRUCTURAL_VARIANT',\n '')\n),\nprofiled AS (\n SELECT sct.cancer_type,\n COUNT(DISTINCT p.sample_unique_id) AS profiled_samples\n FROM profiled_samples_for_gene p\n JOIN cohort c USING (cancer_study_identifier)\n JOIN sample_cancer_type sct USING (sample_unique_id)\n GROUP BY sct.cancer_type\n)\nSELECT a.cancer_type,\n a.altered_samples,\n p.profiled_samples,\n ROUND(a.altered_samples * 100.0 / NULLIF(p.profiled_samples, 0), 1) AS frequency_pct\nFROM altered a\nJOIN profiled p USING (cancer_type)\nWHERE p.profiled_samples >= 50" }, { "name": "154-frequency-msk_chord_2024", "source": "pr154-fallback", "sql": "WITH\n events AS (\n SELECT sample_unique_id,\n (variant_type = 'mutation' AND mutation_status != 'UNCALLED') AS is_mutation,\n (variant_type = 'cna' AND cna_alteration = 2) AS is_amplification,\n (variant_type = 'cna' AND cna_alteration = -2) AS is_deep_deletion,\n (variant_type = 'structural_variant' AND mutation_status != 'UNCALLED') AS is_structural_variant\n FROM genomic_event_derived\n WHERE cancer_study_identifier IN ('msk_chord_2024')\n AND hugo_gene_symbol IN ('TP53')\n AND off_panel = 0\n ),\n profiled AS (\n SELECT stgp.sample_unique_id AS sample_unique_id,\n stgp.alteration_type AS alteration_type\n FROM sample_to_gene_panel_derived stgp\n JOIN gene_panel gp ON stgp.gene_panel_id = gp.stable_id\n JOIN gene_panel_list gpl ON gp.internal_id = gpl.internal_id\n JOIN gene ge ON gpl.gene_id = ge.entrez_gene_id\n WHERE stgp.cancer_study_identifier IN ('msk_chord_2024')\n AND ge.hugo_gene_symbol IN ('TP53')\n AND (stgp.alteration_type IN ('MUTATION_EXTENDED', 'STRUCTURAL_VARIANT') OR (stgp.alteration_type = 'COPY_NUMBER_ALTERATION' AND stgp.genetic_profile_id IN (SELECT stable_id FROM genetic_profile WHERE datatype = 'DISCRETE')))\n UNION ALL\n SELECT sample_unique_id, alteration_type\n FROM sample_to_gene_panel_derived\n WHERE cancer_study_identifier IN ('msk_chord_2024')\n AND gene_panel_id = 'WES'\n AND (alteration_type IN ('MUTATION_EXTENDED', 'STRUCTURAL_VARIANT') OR (alteration_type = 'COPY_NUMBER_ALTERATION' AND genetic_profile_id IN (SELECT stable_id FROM genetic_profile WHERE datatype = 'DISCRETE')))\n )\n SELECT *\n FROM (\n SELECT groupUniqArray(hugo_gene_symbol) AS matched_genes\n FROM gene WHERE hugo_gene_symbol IN ('TP53')\n ) AS gene_match\n CROSS JOIN (\n SELECT groupUniqArray(cancer_study_identifier) AS matched_studies\n FROM cancer_study WHERE cancer_study_identifier IN ('msk_chord_2024')\n ) AS study_match\n CROSS JOIN (\n SELECT uniqExactIf(sample_unique_id, is_mutation) AS altered_mutation,\n uniqExactIf(sample_unique_id, is_amplification) AS altered_amplification,\n uniqExactIf(sample_unique_id, is_deep_deletion) AS altered_deep_deletion,\n uniqExactIf(sample_unique_id, is_structural_variant)\n AS altered_structural_variant,\n uniqExactIf(sample_unique_id, is_mutation OR is_amplification\n OR is_deep_deletion OR is_structural_variant) AS altered_any\n FROM events\n ) AS altered\n CROSS JOIN (\n SELECT uniqExactIf(sample_unique_id, alteration_type = 'MUTATION_EXTENDED')\n AS profiled_MUTATION_EXTENDED,\n uniqExactIf(sample_unique_id, alteration_type = 'COPY_NUMBER_ALTERATION')\n AS profiled_COPY_NUMBER_ALTERATION,\n uniqExactIf(sample_unique_id, alteration_type = 'STRUCTURAL_VARIANT')\n AS profiled_STRUCTURAL_VARIANT,\n uniqExact(sample_unique_id) AS profiled_ANY\n FROM profiled\n ) AS profiled_counts" }, { "name": "154-top-mutation-msk_chord_2024", "source": "pr154-fallback", "sql": "WITH\n altered AS (\n SELECT hugo_gene_symbol, COUNT(DISTINCT sample_unique_id) AS altered_samples\n FROM genomic_event_derived\n WHERE cancer_study_identifier IN ('msk_chord_2024')\n AND off_panel = 0\n AND (variant_type = 'mutation' AND mutation_status != 'UNCALLED')\n GROUP BY hugo_gene_symbol\n ORDER BY altered_samples DESC, hugo_gene_symbol ASC\n LIMIT 20\n ),\n wes_samples AS (\n SELECT DISTINCT sample_unique_id\n FROM sample_to_gene_panel_derived\n WHERE cancer_study_identifier IN ('msk_chord_2024')\n AND gene_panel_id = 'WES'\n AND (alteration_type IN ('MUTATION_EXTENDED'))\n ),\n panel_profiled AS (\n SELECT ge.hugo_gene_symbol AS hugo_gene_symbol,\n COUNT(DISTINCT stgp.sample_unique_id) AS n\n FROM sample_to_gene_panel_derived stgp\n JOIN gene_panel gp ON stgp.gene_panel_id = gp.stable_id\n JOIN gene_panel_list gpl ON gp.internal_id = gpl.internal_id\n JOIN gene ge ON gpl.gene_id = ge.entrez_gene_id\n WHERE stgp.cancer_study_identifier IN ('msk_chord_2024')\n AND (stgp.alteration_type IN ('MUTATION_EXTENDED'))\n AND stgp.sample_unique_id NOT IN (SELECT sample_unique_id FROM wes_samples)\n AND ge.hugo_gene_symbol IN (SELECT hugo_gene_symbol FROM altered)\n GROUP BY ge.hugo_gene_symbol\n )\n SELECT a.hugo_gene_symbol AS hugo_gene_symbol,\n a.altered_samples AS altered_samples,\n (SELECT count() FROM wes_samples) + COALESCE(p.n, 0) AS profiled_samples\n FROM altered a\n LEFT JOIN panel_profiled p ON a.hugo_gene_symbol = p.hugo_gene_symbol\n ORDER BY altered_samples DESC, hugo_gene_symbol ASC" }, { "name": "154-profiled-msk_chord_2024", "source": "pr154-fallback", "sql": "WITH\n profiled AS (\n SELECT cancer_study_identifier, alteration_type AS profile_type,\n sample_unique_id, gene_panel_id\n FROM sample_to_gene_panel_derived\n WHERE cancer_study_identifier IN ('msk_chord_2024')\n UNION ALL\n SELECT cancer_study_identifier, 'COPY_NUMBER_ALTERATION_DISCRETE' AS profile_type,\n sample_unique_id, gene_panel_id\n FROM sample_to_gene_panel_derived\n WHERE cancer_study_identifier IN ('msk_chord_2024')\n AND ((alteration_type = 'COPY_NUMBER_ALTERATION' AND genetic_profile_id IN (SELECT stable_id FROM genetic_profile WHERE datatype = 'DISCRETE')))\n UNION ALL\n SELECT cancer_study_identifier, 'ANY_MUT_CNA_SV' AS profile_type,\n sample_unique_id, gene_panel_id\n FROM sample_to_gene_panel_derived\n WHERE cancer_study_identifier IN ('msk_chord_2024')\n AND (alteration_type IN ('MUTATION_EXTENDED', 'STRUCTURAL_VARIANT') OR (alteration_type = 'COPY_NUMBER_ALTERATION' AND genetic_profile_id IN (SELECT stable_id FROM genetic_profile WHERE datatype = 'DISCRETE')))\n ),\n sample_patient AS (\n SELECT sample_unique_id, patient_unique_id\n FROM sample_derived\n WHERE cancer_study_identifier IN ('msk_chord_2024')\n )\n SELECT * FROM (\n SELECT cancer_study_identifier, 'ALL_SAMPLES' AS profile_type,\n COUNT(DISTINCT sample_unique_id) AS samples,\n COUNT(DISTINCT patient_unique_id) AS patients,\n toUInt64(0) AS wes_samples\n FROM sample_derived\n WHERE cancer_study_identifier IN ('msk_chord_2024')\n GROUP BY cancer_study_identifier\n UNION ALL\n SELECT p.cancer_study_identifier, p.profile_type,\n COUNT(DISTINCT p.sample_unique_id),\n COUNT(DISTINCT nullIf(sp.patient_unique_id, '')),\n COUNT(DISTINCT if(p.gene_panel_id = 'WES', p.sample_unique_id, NULL))\n FROM profiled p\n LEFT JOIN sample_patient sp ON p.sample_unique_id = sp.sample_unique_id\n GROUP BY p.cancer_study_identifier, p.profile_type\n )\n ORDER BY cancer_study_identifier, profile_type = 'ALL_SAMPLES' DESC,\n profile_type = 'ANY_MUT_CNA_SV' DESC, samples DESC, profile_type" }, { "name": "154-frequency-brca_tcga_pan_can_atlas_2018", "source": "pr154-fallback", "sql": "WITH\n events AS (\n SELECT sample_unique_id,\n (variant_type = 'mutation' AND mutation_status != 'UNCALLED') AS is_mutation,\n (variant_type = 'cna' AND cna_alteration = 2) AS is_amplification,\n (variant_type = 'cna' AND cna_alteration = -2) AS is_deep_deletion,\n (variant_type = 'structural_variant' AND mutation_status != 'UNCALLED') AS is_structural_variant\n FROM genomic_event_derived\n WHERE cancer_study_identifier IN ('brca_tcga_pan_can_atlas_2018')\n AND hugo_gene_symbol IN ('TP53')\n AND off_panel = 0\n ),\n profiled AS (\n SELECT stgp.sample_unique_id AS sample_unique_id,\n stgp.alteration_type AS alteration_type\n FROM sample_to_gene_panel_derived stgp\n JOIN gene_panel gp ON stgp.gene_panel_id = gp.stable_id\n JOIN gene_panel_list gpl ON gp.internal_id = gpl.internal_id\n JOIN gene ge ON gpl.gene_id = ge.entrez_gene_id\n WHERE stgp.cancer_study_identifier IN ('brca_tcga_pan_can_atlas_2018')\n AND ge.hugo_gene_symbol IN ('TP53')\n AND (stgp.alteration_type IN ('MUTATION_EXTENDED', 'STRUCTURAL_VARIANT') OR (stgp.alteration_type = 'COPY_NUMBER_ALTERATION' AND stgp.genetic_profile_id IN (SELECT stable_id FROM genetic_profile WHERE datatype = 'DISCRETE')))\n UNION ALL\n SELECT sample_unique_id, alteration_type\n FROM sample_to_gene_panel_derived\n WHERE cancer_study_identifier IN ('brca_tcga_pan_can_atlas_2018')\n AND gene_panel_id = 'WES'\n AND (alteration_type IN ('MUTATION_EXTENDED', 'STRUCTURAL_VARIANT') OR (alteration_type = 'COPY_NUMBER_ALTERATION' AND genetic_profile_id IN (SELECT stable_id FROM genetic_profile WHERE datatype = 'DISCRETE')))\n )\n SELECT *\n FROM (\n SELECT groupUniqArray(hugo_gene_symbol) AS matched_genes\n FROM gene WHERE hugo_gene_symbol IN ('TP53')\n ) AS gene_match\n CROSS JOIN (\n SELECT groupUniqArray(cancer_study_identifier) AS matched_studies\n FROM cancer_study WHERE cancer_study_identifier IN ('brca_tcga_pan_can_atlas_2018')\n ) AS study_match\n CROSS JOIN (\n SELECT uniqExactIf(sample_unique_id, is_mutation) AS altered_mutation,\n uniqExactIf(sample_unique_id, is_amplification) AS altered_amplification,\n uniqExactIf(sample_unique_id, is_deep_deletion) AS altered_deep_deletion,\n uniqExactIf(sample_unique_id, is_structural_variant)\n AS altered_structural_variant,\n uniqExactIf(sample_unique_id, is_mutation OR is_amplification\n OR is_deep_deletion OR is_structural_variant) AS altered_any\n FROM events\n ) AS altered\n CROSS JOIN (\n SELECT uniqExactIf(sample_unique_id, alteration_type = 'MUTATION_EXTENDED')\n AS profiled_MUTATION_EXTENDED,\n uniqExactIf(sample_unique_id, alteration_type = 'COPY_NUMBER_ALTERATION')\n AS profiled_COPY_NUMBER_ALTERATION,\n uniqExactIf(sample_unique_id, alteration_type = 'STRUCTURAL_VARIANT')\n AS profiled_STRUCTURAL_VARIANT,\n uniqExact(sample_unique_id) AS profiled_ANY\n FROM profiled\n ) AS profiled_counts" }, { "name": "154-top-mutation-brca_tcga_pan_can_atlas_2018", "source": "pr154-fallback", "sql": "WITH\n altered AS (\n SELECT hugo_gene_symbol, COUNT(DISTINCT sample_unique_id) AS altered_samples\n FROM genomic_event_derived\n WHERE cancer_study_identifier IN ('brca_tcga_pan_can_atlas_2018')\n AND off_panel = 0\n AND (variant_type = 'mutation' AND mutation_status != 'UNCALLED')\n GROUP BY hugo_gene_symbol\n ORDER BY altered_samples DESC, hugo_gene_symbol ASC\n LIMIT 20\n ),\n wes_samples AS (\n SELECT DISTINCT sample_unique_id\n FROM sample_to_gene_panel_derived\n WHERE cancer_study_identifier IN ('brca_tcga_pan_can_atlas_2018')\n AND gene_panel_id = 'WES'\n AND (alteration_type IN ('MUTATION_EXTENDED'))\n ),\n panel_profiled AS (\n SELECT ge.hugo_gene_symbol AS hugo_gene_symbol,\n COUNT(DISTINCT stgp.sample_unique_id) AS n\n FROM sample_to_gene_panel_derived stgp\n JOIN gene_panel gp ON stgp.gene_panel_id = gp.stable_id\n JOIN gene_panel_list gpl ON gp.internal_id = gpl.internal_id\n JOIN gene ge ON gpl.gene_id = ge.entrez_gene_id\n WHERE stgp.cancer_study_identifier IN ('brca_tcga_pan_can_atlas_2018')\n AND (stgp.alteration_type IN ('MUTATION_EXTENDED'))\n AND stgp.sample_unique_id NOT IN (SELECT sample_unique_id FROM wes_samples)\n AND ge.hugo_gene_symbol IN (SELECT hugo_gene_symbol FROM altered)\n GROUP BY ge.hugo_gene_symbol\n )\n SELECT a.hugo_gene_symbol AS hugo_gene_symbol,\n a.altered_samples AS altered_samples,\n (SELECT count() FROM wes_samples) + COALESCE(p.n, 0) AS profiled_samples\n FROM altered a\n LEFT JOIN panel_profiled p ON a.hugo_gene_symbol = p.hugo_gene_symbol\n ORDER BY altered_samples DESC, hugo_gene_symbol ASC" }, { "name": "154-profiled-brca_tcga_pan_can_atlas_2018", "source": "pr154-fallback", "sql": "WITH\n profiled AS (\n SELECT cancer_study_identifier, alteration_type AS profile_type,\n sample_unique_id, gene_panel_id\n FROM sample_to_gene_panel_derived\n WHERE cancer_study_identifier IN ('brca_tcga_pan_can_atlas_2018')\n UNION ALL\n SELECT cancer_study_identifier, 'COPY_NUMBER_ALTERATION_DISCRETE' AS profile_type,\n sample_unique_id, gene_panel_id\n FROM sample_to_gene_panel_derived\n WHERE cancer_study_identifier IN ('brca_tcga_pan_can_atlas_2018')\n AND ((alteration_type = 'COPY_NUMBER_ALTERATION' AND genetic_profile_id IN (SELECT stable_id FROM genetic_profile WHERE datatype = 'DISCRETE')))\n UNION ALL\n SELECT cancer_study_identifier, 'ANY_MUT_CNA_SV' AS profile_type,\n sample_unique_id, gene_panel_id\n FROM sample_to_gene_panel_derived\n WHERE cancer_study_identifier IN ('brca_tcga_pan_can_atlas_2018')\n AND (alteration_type IN ('MUTATION_EXTENDED', 'STRUCTURAL_VARIANT') OR (alteration_type = 'COPY_NUMBER_ALTERATION' AND genetic_profile_id IN (SELECT stable_id FROM genetic_profile WHERE datatype = 'DISCRETE')))\n ),\n sample_patient AS (\n SELECT sample_unique_id, patient_unique_id\n FROM sample_derived\n WHERE cancer_study_identifier IN ('brca_tcga_pan_can_atlas_2018')\n )\n SELECT * FROM (\n SELECT cancer_study_identifier, 'ALL_SAMPLES' AS profile_type,\n COUNT(DISTINCT sample_unique_id) AS samples,\n COUNT(DISTINCT patient_unique_id) AS patients,\n toUInt64(0) AS wes_samples\n FROM sample_derived\n WHERE cancer_study_identifier IN ('brca_tcga_pan_can_atlas_2018')\n GROUP BY cancer_study_identifier\n UNION ALL\n SELECT p.cancer_study_identifier, p.profile_type,\n COUNT(DISTINCT p.sample_unique_id),\n COUNT(DISTINCT nullIf(sp.patient_unique_id, '')),\n COUNT(DISTINCT if(p.gene_panel_id = 'WES', p.sample_unique_id, NULL))\n FROM profiled p\n LEFT JOIN sample_patient sp ON p.sample_unique_id = sp.sample_unique_id\n GROUP BY p.cancer_study_identifier, p.profile_type\n )\n ORDER BY cancer_study_identifier, profile_type = 'ALL_SAMPLES' DESC,\n profile_type = 'ANY_MUT_CNA_SV' DESC, samples DESC, profile_type" }, { "name": "154-cancer-type-TP53", "source": "pr154-fallback", "sql": "SELECT * FROM (\n SELECT cancer_type, 'TP53' AS hugo_gene_symbol,\n altered_samples, profiled_samples\n FROM gene_alteration_frequency_by_cancer_type(\n preference='all_studies_non_redundant', gene='TP53',\n alteration='mutation')\n )\n ORDER BY altered_samples / profiled_samples DESC, altered_samples DESC, cancer_type ASC\n LIMIT 20" }, { "name": "154-cancer-type-TP53-pan-cancer-tcga", "source": "pr154-fallback", "sql": "SELECT * FROM (\n SELECT cancer_type, 'TP53' AS hugo_gene_symbol,\n altered_samples, profiled_samples\n FROM gene_alteration_frequency_by_cancer_type(\n preference='pan_cancer_tcga', gene='TP53',\n alteration='mutation')\n )\n ORDER BY altered_samples / profiled_samples DESC, altered_samples DESC, cancer_type ASC\n LIMIT 20" }, { "name": "154-build-study_gene_alteration_counts-msk_chord_2024", "source": "pr154-builds", "sql": "WITH\nalteration_to_profile AS (\n SELECT t.1 AS alteration_type, t.2 AS profile_type\n FROM (\n SELECT arrayJoin([\n ('mutation', 'MUTATION_EXTENDED'),\n ('amplification', 'COPY_NUMBER_ALTERATION'),\n ('deep_deletion', 'COPY_NUMBER_ALTERATION'),\n ('structural_variant', 'STRUCTURAL_VARIANT'),\n ('any', 'ANY')\n ]) AS t\n )\n),\nevents AS (\n SELECT cancer_study_identifier,\n hugo_gene_symbol,\n sample_unique_id,\n multiIf(\n variant_type = 'mutation' AND mutation_status != 'UNCALLED', 'mutation',\n variant_type = 'cna' AND cna_alteration = 2, 'amplification',\n variant_type = 'cna' AND cna_alteration = -2, 'deep_deletion',\n variant_type = 'structural_variant' AND mutation_status != 'UNCALLED',\n 'structural_variant',\n '') AS event_type\n FROM (SELECT * FROM genomic_event_derived WHERE cancer_study_identifier='msk_chord_2024')\n WHERE off_panel = 0\n),\naltered AS (\n -- event_type, not alteration_type: in ClickHouse a SELECT alias shadows\n -- the source column in WHERE, so `'any' AS alteration_type` below would\n -- turn a WHERE on alteration_type into a constant.\n SELECT cancer_study_identifier, hugo_gene_symbol, event_type AS alteration_type,\n COUNT(DISTINCT sample_unique_id) AS altered_samples,\n COUNT(*) AS altered_events\n FROM events\n WHERE event_type != ''\n GROUP BY cancer_study_identifier, hugo_gene_symbol, event_type\n UNION ALL\n SELECT cancer_study_identifier, hugo_gene_symbol, 'any' AS alteration_type,\n COUNT(DISTINCT sample_unique_id) AS altered_samples,\n COUNT(*) AS altered_events\n FROM events\n WHERE event_type != ''\n GROUP BY cancer_study_identifier, hugo_gene_symbol\n),\ntyped_profile_rows AS (\n -- Same profile filters as the {mutation,cna,sv}_*_coverage views:\n -- CNA rows only from DISCRETE profiles.\n SELECT cancer_study_identifier, alteration_type AS profile_type, sample_unique_id, gene_panel_id\n FROM (SELECT * FROM sample_to_gene_panel_derived WHERE cancer_study_identifier='msk_chord_2024')\n WHERE alteration_type IN ('MUTATION_EXTENDED', 'STRUCTURAL_VARIANT')\n OR (alteration_type = 'COPY_NUMBER_ALTERATION'\n AND genetic_profile_id IN (SELECT stable_id FROM genetic_profile WHERE datatype = 'DISCRETE'))\n),\nprofile_rows AS (\n SELECT cancer_study_identifier, profile_type, sample_unique_id, gene_panel_id\n FROM typed_profile_rows\n UNION ALL\n SELECT cancer_study_identifier, 'ANY' AS profile_type, sample_unique_id, gene_panel_id\n FROM typed_profile_rows\n),\nsample_buckets AS (\n -- One row per sample per profile type: its panel-set signature.\n SELECT cancer_study_identifier, profile_type, sample_unique_id,\n arraySort(groupUniqArray(gene_panel_id)) AS panels\n FROM profile_rows\n GROUP BY cancer_study_identifier, profile_type, sample_unique_id\n),\nwes_profiled AS (\n SELECT cancer_study_identifier, profile_type, COUNT(*) AS n\n FROM sample_buckets\n WHERE has(panels, 'WES')\n GROUP BY cancer_study_identifier, profile_type\n),\nsignature_counts AS (\n SELECT cancer_study_identifier, profile_type, panels, COUNT(*) AS n\n FROM sample_buckets\n WHERE NOT has(panels, 'WES')\n GROUP BY cancer_study_identifier, profile_type, panels\n),\npanel_genes AS (\n -- Same panel -> gene chain as mutation_panel_gene_coverage.\n SELECT gp.stable_id AS gene_panel_id, g.hugo_gene_symbol AS hugo_gene_symbol\n FROM gene_panel gp\n JOIN gene_panel_list gpl ON gp.internal_id = gpl.internal_id\n JOIN gene g ON gpl.gene_id = g.entrez_gene_id\n),\npanel_profiled AS (\n SELECT cancer_study_identifier, profile_type, hugo_gene_symbol, SUM(n) AS n\n FROM (\n -- DISTINCT: a gene listed by two panels of one signature counts once.\n SELECT DISTINCT s.cancer_study_identifier, s.profile_type, s.panels, s.n, pg.hugo_gene_symbol\n FROM (\n SELECT cancer_study_identifier, profile_type, panels, n,\n arrayJoin(panels) AS gene_panel_id\n FROM signature_counts\n ) s\n JOIN panel_genes pg ON s.gene_panel_id = pg.gene_panel_id\n )\n GROUP BY cancer_study_identifier, profile_type, hugo_gene_symbol\n)\nSELECT a.cancer_study_identifier,\n a.hugo_gene_symbol,\n a.alteration_type,\n a.altered_samples,\n COALESCE(w.n, 0) + COALESCE(p.n, 0) AS profiled_samples,\n a.altered_events,\n now() AS built_at\nFROM altered a\nJOIN alteration_to_profile m ON a.alteration_type = m.alteration_type\nLEFT JOIN wes_profiled w\n ON w.cancer_study_identifier = a.cancer_study_identifier\n AND w.profile_type = m.profile_type\nLEFT JOIN panel_profiled p\n ON p.cancer_study_identifier = a.cancer_study_identifier\n AND p.profile_type = m.profile_type\n AND p.hugo_gene_symbol = a.hugo_gene_symbol" }, { "name": "154-build-study_gene_alteration_counts-brca_tcga_pan_can_atlas_2018", "source": "pr154-builds", "sql": "WITH\nalteration_to_profile AS (\n SELECT t.1 AS alteration_type, t.2 AS profile_type\n FROM (\n SELECT arrayJoin([\n ('mutation', 'MUTATION_EXTENDED'),\n ('amplification', 'COPY_NUMBER_ALTERATION'),\n ('deep_deletion', 'COPY_NUMBER_ALTERATION'),\n ('structural_variant', 'STRUCTURAL_VARIANT'),\n ('any', 'ANY')\n ]) AS t\n )\n),\nevents AS (\n SELECT cancer_study_identifier,\n hugo_gene_symbol,\n sample_unique_id,\n multiIf(\n variant_type = 'mutation' AND mutation_status != 'UNCALLED', 'mutation',\n variant_type = 'cna' AND cna_alteration = 2, 'amplification',\n variant_type = 'cna' AND cna_alteration = -2, 'deep_deletion',\n variant_type = 'structural_variant' AND mutation_status != 'UNCALLED',\n 'structural_variant',\n '') AS event_type\n FROM (SELECT * FROM genomic_event_derived WHERE cancer_study_identifier='brca_tcga_pan_can_atlas_2018')\n WHERE off_panel = 0\n),\naltered AS (\n -- event_type, not alteration_type: in ClickHouse a SELECT alias shadows\n -- the source column in WHERE, so `'any' AS alteration_type` below would\n -- turn a WHERE on alteration_type into a constant.\n SELECT cancer_study_identifier, hugo_gene_symbol, event_type AS alteration_type,\n COUNT(DISTINCT sample_unique_id) AS altered_samples,\n COUNT(*) AS altered_events\n FROM events\n WHERE event_type != ''\n GROUP BY cancer_study_identifier, hugo_gene_symbol, event_type\n UNION ALL\n SELECT cancer_study_identifier, hugo_gene_symbol, 'any' AS alteration_type,\n COUNT(DISTINCT sample_unique_id) AS altered_samples,\n COUNT(*) AS altered_events\n FROM events\n WHERE event_type != ''\n GROUP BY cancer_study_identifier, hugo_gene_symbol\n),\ntyped_profile_rows AS (\n -- Same profile filters as the {mutation,cna,sv}_*_coverage views:\n -- CNA rows only from DISCRETE profiles.\n SELECT cancer_study_identifier, alteration_type AS profile_type, sample_unique_id, gene_panel_id\n FROM (SELECT * FROM sample_to_gene_panel_derived WHERE cancer_study_identifier='brca_tcga_pan_can_atlas_2018')\n WHERE alteration_type IN ('MUTATION_EXTENDED', 'STRUCTURAL_VARIANT')\n OR (alteration_type = 'COPY_NUMBER_ALTERATION'\n AND genetic_profile_id IN (SELECT stable_id FROM genetic_profile WHERE datatype = 'DISCRETE'))\n),\nprofile_rows AS (\n SELECT cancer_study_identifier, profile_type, sample_unique_id, gene_panel_id\n FROM typed_profile_rows\n UNION ALL\n SELECT cancer_study_identifier, 'ANY' AS profile_type, sample_unique_id, gene_panel_id\n FROM typed_profile_rows\n),\nsample_buckets AS (\n -- One row per sample per profile type: its panel-set signature.\n SELECT cancer_study_identifier, profile_type, sample_unique_id,\n arraySort(groupUniqArray(gene_panel_id)) AS panels\n FROM profile_rows\n GROUP BY cancer_study_identifier, profile_type, sample_unique_id\n),\nwes_profiled AS (\n SELECT cancer_study_identifier, profile_type, COUNT(*) AS n\n FROM sample_buckets\n WHERE has(panels, 'WES')\n GROUP BY cancer_study_identifier, profile_type\n),\nsignature_counts AS (\n SELECT cancer_study_identifier, profile_type, panels, COUNT(*) AS n\n FROM sample_buckets\n WHERE NOT has(panels, 'WES')\n GROUP BY cancer_study_identifier, profile_type, panels\n),\npanel_genes AS (\n -- Same panel -> gene chain as mutation_panel_gene_coverage.\n SELECT gp.stable_id AS gene_panel_id, g.hugo_gene_symbol AS hugo_gene_symbol\n FROM gene_panel gp\n JOIN gene_panel_list gpl ON gp.internal_id = gpl.internal_id\n JOIN gene g ON gpl.gene_id = g.entrez_gene_id\n),\npanel_profiled AS (\n SELECT cancer_study_identifier, profile_type, hugo_gene_symbol, SUM(n) AS n\n FROM (\n -- DISTINCT: a gene listed by two panels of one signature counts once.\n SELECT DISTINCT s.cancer_study_identifier, s.profile_type, s.panels, s.n, pg.hugo_gene_symbol\n FROM (\n SELECT cancer_study_identifier, profile_type, panels, n,\n arrayJoin(panels) AS gene_panel_id\n FROM signature_counts\n ) s\n JOIN panel_genes pg ON s.gene_panel_id = pg.gene_panel_id\n )\n GROUP BY cancer_study_identifier, profile_type, hugo_gene_symbol\n)\nSELECT a.cancer_study_identifier,\n a.hugo_gene_symbol,\n a.alteration_type,\n a.altered_samples,\n COALESCE(w.n, 0) + COALESCE(p.n, 0) AS profiled_samples,\n a.altered_events,\n now() AS built_at\nFROM altered a\nJOIN alteration_to_profile m ON a.alteration_type = m.alteration_type\nLEFT JOIN wes_profiled w\n ON w.cancer_study_identifier = a.cancer_study_identifier\n AND w.profile_type = m.profile_type\nLEFT JOIN panel_profiled p\n ON p.cancer_study_identifier = a.cancer_study_identifier\n AND p.profile_type = m.profile_type\n AND p.hugo_gene_symbol = a.hugo_gene_symbol" }, { "name": "154-build-cancer_type_gene_alteration_counts-msk_chord_2024", "source": "pr154-builds", "sql": "WITH\nalteration_to_profile AS (\n SELECT t.1 AS alteration_type, t.2 AS profile_type\n FROM (\n SELECT arrayJoin([\n ('mutation', 'MUTATION_EXTENDED'),\n ('amplification', 'COPY_NUMBER_ALTERATION'),\n ('deep_deletion', 'COPY_NUMBER_ALTERATION'),\n ('structural_variant', 'STRUCTURAL_VARIANT'),\n ('any', 'ANY')\n ]) AS t\n )\n),\nsample_cancer_type AS (\n SELECT q.preference_name AS preference_name,\n cd.cancer_study_identifier AS cancer_study_identifier,\n cd.sample_unique_id AS sample_unique_id,\n cd.attribute_value AS cancer_type\n FROM (SELECT * FROM clinical_data_derived WHERE cancer_study_identifier='msk_chord_2024') cd\n JOIN cancer_study_query_preferences q ON cd.cancer_study_identifier = q.cancer_study_identifier\n WHERE cd.attribute_name = 'CANCER_TYPE'\n),\nevents AS (\n SELECT cancer_study_identifier,\n hugo_gene_symbol,\n sample_unique_id,\n multiIf(\n variant_type = 'mutation' AND mutation_status != 'UNCALLED', 'mutation',\n variant_type = 'cna' AND cna_alteration = 2, 'amplification',\n variant_type = 'cna' AND cna_alteration = -2, 'deep_deletion',\n variant_type = 'structural_variant' AND mutation_status != 'UNCALLED',\n 'structural_variant',\n '') AS event_type\n FROM (SELECT * FROM genomic_event_derived WHERE cancer_study_identifier='msk_chord_2024')\n WHERE off_panel = 0\n AND cancer_study_identifier IN (SELECT cancer_study_identifier FROM cancer_study_query_preferences)\n),\ntyped_events AS (\n SELECT sct.preference_name AS preference_name, sct.cancer_type AS cancer_type,\n e.hugo_gene_symbol AS hugo_gene_symbol, e.event_type AS event_type,\n e.sample_unique_id AS sample_unique_id\n FROM events e\n JOIN sample_cancer_type sct\n ON e.cancer_study_identifier = sct.cancer_study_identifier\n AND e.sample_unique_id = sct.sample_unique_id\n WHERE e.event_type != ''\n),\naltered AS (\n SELECT preference_name, cancer_type, hugo_gene_symbol, event_type AS alteration_type,\n COUNT(DISTINCT sample_unique_id) AS altered_samples\n FROM typed_events\n GROUP BY preference_name, cancer_type, hugo_gene_symbol, event_type\n UNION ALL\n SELECT preference_name, cancer_type, hugo_gene_symbol, 'any' AS alteration_type,\n COUNT(DISTINCT sample_unique_id) AS altered_samples\n FROM typed_events\n GROUP BY preference_name, cancer_type, hugo_gene_symbol\n),\nprofile_rows AS (\n SELECT cancer_study_identifier, alteration_type AS profile_type, sample_unique_id, gene_panel_id\n FROM (SELECT * FROM sample_to_gene_panel_derived WHERE cancer_study_identifier='msk_chord_2024')\n WHERE alteration_type IN ('MUTATION_EXTENDED', 'COPY_NUMBER_ALTERATION', 'STRUCTURAL_VARIANT')\n UNION ALL\n SELECT cancer_study_identifier, 'ANY' AS profile_type, sample_unique_id, gene_panel_id\n FROM (SELECT * FROM sample_to_gene_panel_derived WHERE cancer_study_identifier='msk_chord_2024')\n WHERE alteration_type IN ('MUTATION_EXTENDED', 'COPY_NUMBER_ALTERATION', 'STRUCTURAL_VARIANT')\n),\nsample_buckets AS (\n SELECT sct.preference_name AS preference_name, sct.cancer_type AS cancer_type,\n pr.profile_type AS profile_type, pr.sample_unique_id AS sample_unique_id,\n arraySort(groupUniqArray(pr.gene_panel_id)) AS panels\n FROM profile_rows pr\n JOIN sample_cancer_type sct\n ON pr.cancer_study_identifier = sct.cancer_study_identifier\n AND pr.sample_unique_id = sct.sample_unique_id\n GROUP BY preference_name, cancer_type, profile_type, sample_unique_id\n),\nwes_profiled AS (\n SELECT preference_name, cancer_type, profile_type, COUNT(*) AS n\n FROM sample_buckets\n WHERE has(panels, 'WES')\n GROUP BY preference_name, cancer_type, profile_type\n),\nsignature_counts AS (\n SELECT preference_name, cancer_type, profile_type, panels, COUNT(*) AS n\n FROM sample_buckets\n WHERE NOT has(panels, 'WES')\n GROUP BY preference_name, cancer_type, profile_type, panels\n),\npanel_genes AS (\n SELECT gp.stable_id AS gene_panel_id, g.hugo_gene_symbol AS hugo_gene_symbol\n FROM gene_panel gp\n JOIN gene_panel_list gpl ON gp.internal_id = gpl.internal_id\n JOIN gene g ON gpl.gene_id = g.entrez_gene_id\n),\npanel_profiled AS (\n SELECT preference_name, cancer_type, profile_type, hugo_gene_symbol, SUM(n) AS n\n FROM (\n SELECT DISTINCT s.preference_name, s.cancer_type, s.profile_type, s.panels, s.n,\n pg.hugo_gene_symbol\n FROM (\n SELECT preference_name, cancer_type, profile_type, panels, n,\n arrayJoin(panels) AS gene_panel_id\n FROM signature_counts\n ) s\n JOIN panel_genes pg ON s.gene_panel_id = pg.gene_panel_id\n )\n GROUP BY preference_name, cancer_type, profile_type, hugo_gene_symbol\n)\nSELECT a.preference_name,\n a.cancer_type,\n a.hugo_gene_symbol,\n a.alteration_type,\n a.altered_samples,\n COALESCE(w.n, 0) + COALESCE(p.n, 0) AS profiled_samples,\n now() AS built_at\nFROM altered a\nJOIN alteration_to_profile m ON a.alteration_type = m.alteration_type\nLEFT JOIN wes_profiled w\n ON w.preference_name = a.preference_name\n AND w.cancer_type = a.cancer_type\n AND w.profile_type = m.profile_type\nLEFT JOIN panel_profiled p\n ON p.preference_name = a.preference_name\n AND p.cancer_type = a.cancer_type\n AND p.profile_type = m.profile_type\n AND p.hugo_gene_symbol = a.hugo_gene_symbol" }, { "name": "154-build-cancer_type_gene_alteration_counts-brca_tcga_pan_can_atlas_2018", "source": "pr154-builds", "sql": "WITH\nalteration_to_profile AS (\n SELECT t.1 AS alteration_type, t.2 AS profile_type\n FROM (\n SELECT arrayJoin([\n ('mutation', 'MUTATION_EXTENDED'),\n ('amplification', 'COPY_NUMBER_ALTERATION'),\n ('deep_deletion', 'COPY_NUMBER_ALTERATION'),\n ('structural_variant', 'STRUCTURAL_VARIANT'),\n ('any', 'ANY')\n ]) AS t\n )\n),\nsample_cancer_type AS (\n SELECT q.preference_name AS preference_name,\n cd.cancer_study_identifier AS cancer_study_identifier,\n cd.sample_unique_id AS sample_unique_id,\n cd.attribute_value AS cancer_type\n FROM (SELECT * FROM clinical_data_derived WHERE cancer_study_identifier='brca_tcga_pan_can_atlas_2018') cd\n JOIN cancer_study_query_preferences q ON cd.cancer_study_identifier = q.cancer_study_identifier\n WHERE cd.attribute_name = 'CANCER_TYPE'\n),\nevents AS (\n SELECT cancer_study_identifier,\n hugo_gene_symbol,\n sample_unique_id,\n multiIf(\n variant_type = 'mutation' AND mutation_status != 'UNCALLED', 'mutation',\n variant_type = 'cna' AND cna_alteration = 2, 'amplification',\n variant_type = 'cna' AND cna_alteration = -2, 'deep_deletion',\n variant_type = 'structural_variant' AND mutation_status != 'UNCALLED',\n 'structural_variant',\n '') AS event_type\n FROM (SELECT * FROM genomic_event_derived WHERE cancer_study_identifier='brca_tcga_pan_can_atlas_2018')\n WHERE off_panel = 0\n AND cancer_study_identifier IN (SELECT cancer_study_identifier FROM cancer_study_query_preferences)\n),\ntyped_events AS (\n SELECT sct.preference_name AS preference_name, sct.cancer_type AS cancer_type,\n e.hugo_gene_symbol AS hugo_gene_symbol, e.event_type AS event_type,\n e.sample_unique_id AS sample_unique_id\n FROM events e\n JOIN sample_cancer_type sct\n ON e.cancer_study_identifier = sct.cancer_study_identifier\n AND e.sample_unique_id = sct.sample_unique_id\n WHERE e.event_type != ''\n),\naltered AS (\n SELECT preference_name, cancer_type, hugo_gene_symbol, event_type AS alteration_type,\n COUNT(DISTINCT sample_unique_id) AS altered_samples\n FROM typed_events\n GROUP BY preference_name, cancer_type, hugo_gene_symbol, event_type\n UNION ALL\n SELECT preference_name, cancer_type, hugo_gene_symbol, 'any' AS alteration_type,\n COUNT(DISTINCT sample_unique_id) AS altered_samples\n FROM typed_events\n GROUP BY preference_name, cancer_type, hugo_gene_symbol\n),\nprofile_rows AS (\n SELECT cancer_study_identifier, alteration_type AS profile_type, sample_unique_id, gene_panel_id\n FROM (SELECT * FROM sample_to_gene_panel_derived WHERE cancer_study_identifier='brca_tcga_pan_can_atlas_2018')\n WHERE alteration_type IN ('MUTATION_EXTENDED', 'COPY_NUMBER_ALTERATION', 'STRUCTURAL_VARIANT')\n UNION ALL\n SELECT cancer_study_identifier, 'ANY' AS profile_type, sample_unique_id, gene_panel_id\n FROM (SELECT * FROM sample_to_gene_panel_derived WHERE cancer_study_identifier='brca_tcga_pan_can_atlas_2018')\n WHERE alteration_type IN ('MUTATION_EXTENDED', 'COPY_NUMBER_ALTERATION', 'STRUCTURAL_VARIANT')\n),\nsample_buckets AS (\n SELECT sct.preference_name AS preference_name, sct.cancer_type AS cancer_type,\n pr.profile_type AS profile_type, pr.sample_unique_id AS sample_unique_id,\n arraySort(groupUniqArray(pr.gene_panel_id)) AS panels\n FROM profile_rows pr\n JOIN sample_cancer_type sct\n ON pr.cancer_study_identifier = sct.cancer_study_identifier\n AND pr.sample_unique_id = sct.sample_unique_id\n GROUP BY preference_name, cancer_type, profile_type, sample_unique_id\n),\nwes_profiled AS (\n SELECT preference_name, cancer_type, profile_type, COUNT(*) AS n\n FROM sample_buckets\n WHERE has(panels, 'WES')\n GROUP BY preference_name, cancer_type, profile_type\n),\nsignature_counts AS (\n SELECT preference_name, cancer_type, profile_type, panels, COUNT(*) AS n\n FROM sample_buckets\n WHERE NOT has(panels, 'WES')\n GROUP BY preference_name, cancer_type, profile_type, panels\n),\npanel_genes AS (\n SELECT gp.stable_id AS gene_panel_id, g.hugo_gene_symbol AS hugo_gene_symbol\n FROM gene_panel gp\n JOIN gene_panel_list gpl ON gp.internal_id = gpl.internal_id\n JOIN gene g ON gpl.gene_id = g.entrez_gene_id\n),\npanel_profiled AS (\n SELECT preference_name, cancer_type, profile_type, hugo_gene_symbol, SUM(n) AS n\n FROM (\n SELECT DISTINCT s.preference_name, s.cancer_type, s.profile_type, s.panels, s.n,\n pg.hugo_gene_symbol\n FROM (\n SELECT preference_name, cancer_type, profile_type, panels, n,\n arrayJoin(panels) AS gene_panel_id\n FROM signature_counts\n ) s\n JOIN panel_genes pg ON s.gene_panel_id = pg.gene_panel_id\n )\n GROUP BY preference_name, cancer_type, profile_type, hugo_gene_symbol\n)\nSELECT a.preference_name,\n a.cancer_type,\n a.hugo_gene_symbol,\n a.alteration_type,\n a.altered_samples,\n COALESCE(w.n, 0) + COALESCE(p.n, 0) AS profiled_samples,\n now() AS built_at\nFROM altered a\nJOIN alteration_to_profile m ON a.alteration_type = m.alteration_type\nLEFT JOIN wes_profiled w\n ON w.preference_name = a.preference_name\n AND w.cancer_type = a.cancer_type\n AND w.profile_type = m.profile_type\nLEFT JOIN panel_profiled p\n ON p.preference_name = a.preference_name\n AND p.cancer_type = a.cancer_type\n AND p.profile_type = m.profile_type\n AND p.hugo_gene_symbol = a.hugo_gene_symbol" }, { "name": "154-build-study_profiled_counts-msk_chord_2024", "source": "pr154-builds", "sql": "WITH\nsample_patient AS (\n SELECT sample_unique_id, patient_unique_id\n FROM (SELECT * FROM sample_derived WHERE cancer_study_identifier='msk_chord_2024')\n),\ndiscrete_cna_profiles AS (\n SELECT stable_id FROM genetic_profile WHERE datatype = 'DISCRETE'\n),\nprofiled AS (\n SELECT cancer_study_identifier, alteration_type AS profile_type, sample_unique_id, gene_panel_id\n FROM (SELECT * FROM sample_to_gene_panel_derived WHERE cancer_study_identifier='msk_chord_2024')\n UNION ALL\n SELECT cancer_study_identifier, 'COPY_NUMBER_ALTERATION_DISCRETE' AS profile_type,\n sample_unique_id, gene_panel_id\n FROM (SELECT * FROM sample_to_gene_panel_derived WHERE cancer_study_identifier='msk_chord_2024')\n WHERE alteration_type = 'COPY_NUMBER_ALTERATION'\n AND genetic_profile_id IN (SELECT stable_id FROM discrete_cna_profiles)\n UNION ALL\n SELECT cancer_study_identifier, 'ANY_MUT_CNA_SV' AS profile_type, sample_unique_id, gene_panel_id\n FROM (SELECT * FROM sample_to_gene_panel_derived WHERE cancer_study_identifier='msk_chord_2024')\n WHERE alteration_type IN ('MUTATION_EXTENDED', 'STRUCTURAL_VARIANT')\n OR (alteration_type = 'COPY_NUMBER_ALTERATION'\n AND genetic_profile_id IN (SELECT stable_id FROM discrete_cna_profiles))\n)\nSELECT cancer_study_identifier,\n 'ALL_SAMPLES' AS profile_type,\n COUNT(DISTINCT sample_unique_id) AS samples,\n COUNT(DISTINCT patient_unique_id) AS patients,\n toUInt64(0) AS wes_samples,\n now() AS built_at\nFROM (SELECT * FROM sample_derived WHERE cancer_study_identifier='msk_chord_2024')\nGROUP BY cancer_study_identifier\nUNION ALL\nSELECT p.cancer_study_identifier,\n p.profile_type,\n COUNT(DISTINCT p.sample_unique_id) AS samples,\n COUNT(DISTINCT nullIf(sp.patient_unique_id, '')) AS patients,\n COUNT(DISTINCT if(p.gene_panel_id = 'WES', p.sample_unique_id, NULL)) AS wes_samples,\n now() AS built_at\nFROM profiled p\nLEFT JOIN sample_patient sp ON p.sample_unique_id = sp.sample_unique_id\nGROUP BY p.cancer_study_identifier, p.profile_type" }, { "name": "154-build-study_profiled_counts-brca_tcga_pan_can_atlas_2018", "source": "pr154-builds", "sql": "WITH\nsample_patient AS (\n SELECT sample_unique_id, patient_unique_id\n FROM (SELECT * FROM sample_derived WHERE cancer_study_identifier='brca_tcga_pan_can_atlas_2018')\n),\ndiscrete_cna_profiles AS (\n SELECT stable_id FROM genetic_profile WHERE datatype = 'DISCRETE'\n),\nprofiled AS (\n SELECT cancer_study_identifier, alteration_type AS profile_type, sample_unique_id, gene_panel_id\n FROM (SELECT * FROM sample_to_gene_panel_derived WHERE cancer_study_identifier='brca_tcga_pan_can_atlas_2018')\n UNION ALL\n SELECT cancer_study_identifier, 'COPY_NUMBER_ALTERATION_DISCRETE' AS profile_type,\n sample_unique_id, gene_panel_id\n FROM (SELECT * FROM sample_to_gene_panel_derived WHERE cancer_study_identifier='brca_tcga_pan_can_atlas_2018')\n WHERE alteration_type = 'COPY_NUMBER_ALTERATION'\n AND genetic_profile_id IN (SELECT stable_id FROM discrete_cna_profiles)\n UNION ALL\n SELECT cancer_study_identifier, 'ANY_MUT_CNA_SV' AS profile_type, sample_unique_id, gene_panel_id\n FROM (SELECT * FROM sample_to_gene_panel_derived WHERE cancer_study_identifier='brca_tcga_pan_can_atlas_2018')\n WHERE alteration_type IN ('MUTATION_EXTENDED', 'STRUCTURAL_VARIANT')\n OR (alteration_type = 'COPY_NUMBER_ALTERATION'\n AND genetic_profile_id IN (SELECT stable_id FROM discrete_cna_profiles))\n)\nSELECT cancer_study_identifier,\n 'ALL_SAMPLES' AS profile_type,\n COUNT(DISTINCT sample_unique_id) AS samples,\n COUNT(DISTINCT patient_unique_id) AS patients,\n toUInt64(0) AS wes_samples,\n now() AS built_at\nFROM (SELECT * FROM sample_derived WHERE cancer_study_identifier='brca_tcga_pan_can_atlas_2018')\nGROUP BY cancer_study_identifier\nUNION ALL\nSELECT p.cancer_study_identifier,\n p.profile_type,\n COUNT(DISTINCT p.sample_unique_id) AS samples,\n COUNT(DISTINCT nullIf(sp.patient_unique_id, '')) AS patients,\n COUNT(DISTINCT if(p.gene_panel_id = 'WES', p.sample_unique_id, NULL)) AS wes_samples,\n now() AS built_at\nFROM profiled p\nLEFT JOIN sample_patient sp ON p.sample_unique_id = sp.sample_unique_id\nGROUP BY p.cancer_study_identifier, p.profile_type" } ]