Release 1.6.7 (#432)

* Fix fusion report text and author name, update test md5 * Fix treatment options text; insert a comma, tidy up spacing, update test md5 * Fix spacing in 'below threshold to call MS score' message, move make_ordinal method into general-purpose HTML builder * Make WGS description past tense throughout * Evaluate and apply thresholds for HRD and MSI; record threshold status in results; plugin tests pass * add total MS sites * Update treatment header text and test md5 * Add description and links for NCCN compendium * Fix TAR description, and move links to versions.py * move microsatellite total into versions.py * update test md5sums * bugfix, MSI cells were not correctly generated * make thresholds for MSI and HRD greater-than-or-equal instead of greater-than in code and disclaimer * delete extra period * Change to FDA/NCCN 'wording * Improved NCCN explanatory text with document versions * Fix NCCN compendium text * update md5 sums * update version and changelog for release 1.6.7
oicr-gsi · Jul 23, 2024 · 184c563 · 184c563
1 parent 79c787f
commit 184c563
Show file tree

Hide file tree

Showing 21 changed files with 120 additions and 74 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,5 +1,11 @@
 # CHANGELOG
 
+## v1.6.7: 2024-07-23
+- GCGI-1396: Fixes to report text requested by clinical geneticist
+- Correct threshold for reporting HRD; genomic landscape plugin has new "sample type" parameter
+- Improved explanation and links for NCCN compendium
+- Added number of MSI sites
+
 ## v1.6.6: 2024-07-19
 - GCGI-1391: Fixed column names in data_CNA_oncoKBgenes_nonDiploid.txt which impacted oncoKB therapy annotation
 

diff --git a/src/lib/djerba/mergers/treatment_options_merger/test/merger_test.py b/src/lib/djerba/mergers/treatment_options_merger/test/merger_test.py
@@ -28,7 +28,7 @@ def test_gene_info(self):
         self.assertEqual(merger.ini_defaults.get(cc.RENDER_PRIORITY), 50)
         html = merger.render(inputs)
         md5_found = self.getMD5_of_string(html)
-        self.assertEqual(md5_found, '07f1328c64a5c19aa3751905af228239')
+        self.assertEqual(md5_found, 'd94877de73bced10b7aeebd8254e8bbc')
 
 if __name__ == '__main__':
     unittest.main()
diff --git a/src/lib/djerba/mergers/treatment_options_merger/treatment_options_template.html b/src/lib/djerba/mergers/treatment_options_merger/treatment_options_template.html
@@ -8,14 +8,10 @@
 <!-- treatment -->
 ${html_builder.section_cells_begin("Treatment Options", True)}
 
-    <p>
-		Review identified <strong>${approved_total}</strong> option(s) indicating an FDA Approved and/or NCCN Recommended Biomarker 
-		<strong>${investig_total}</strong> option(s) indicating investigational therapies,
-		and <strong>${prognostic_total}</strong> option(s) indicating NCCN-listed biomarkers.
-	</p>
+    <p>Review identified <strong>${approved_total}</strong> option(s) indicating an FDA-approved and/or NCCN-compendium listed treatment, <strong>${investig_total}</strong> option(s) indicating investigational therapies, and <strong>${prognostic_total}</strong> option(s) indicating NCCN-listed biomarkers.</p>
 
 	% if approved_total > 0:
-		<h3 class="header3">FDA Approved and/or NCCN Recommended Biomarker:</h3>
+		<h3 class="header3">FDA-approved and/or NCCN-recommended Biomarker:</h3>
 		<table class="variants" width="100%">
 		<thead style="background-color:white">
 		<tr>
@@ -82,4 +78,4 @@ <h3 class="header3">NCCN-listed Biomarker:</h3>
 		</table>
     % endif
 
-${html_builder.section_cells_end()}
+${html_builder.section_cells_end()}
diff --git a/src/lib/djerba/plugins/fusion/fusion_template.html b/src/lib/djerba/plugins/fusion/fusion_template.html
@@ -14,8 +14,7 @@
 
     <p>
       <strong>${results.get(fusion.TOTAL_VARIANTS)}</strong> cancer gene(s) were subject to rearrangement. 
-      <strong>${results.get(fusion.CLINICALLY_RELEVANT_VARIANTS)}</strong> fusion(s) were oncogenic according to OncoKB. 
-      ,in addition to which <strong>${results.get(fusion.NCCN_RELEVANT_VARIANTS)}</strong> rearrangement(s) appeared in NCCN's biomarker compendium.
+      <strong>${results.get(fusion.CLINICALLY_RELEVANT_VARIANTS)}</strong> fusion(s) were oncogenic according to OncoKB and <strong>${results.get(fusion.NCCN_RELEVANT_VARIANTS)}</strong> rearrangement(s) appeared in the NCCN biomarker compendium.
     </p>
 
     % if results.get(fusion.CLINICALLY_RELEVANT_VARIANTS) > 0:

diff --git a/src/lib/djerba/plugins/fusion/test/fusion.ini b/src/lib/djerba/plugins/fusion/test/fusion.ini
@@ -2,7 +2,7 @@
 archive_name = djerba
 archive_url = http://admin:[email protected]:5984
 attributes = 
-author = CGI Author
+author = Test Author
 configure_priority = 100
 depends_configure = 
 depends_extract = 

diff --git a/src/lib/djerba/plugins/fusion/test/plugin_test.py b/src/lib/djerba/plugins/fusion/test/plugin_test.py
@@ -50,7 +50,7 @@ def test(self):
         params = {
             self.INI: self.INI_NAME,
             self.JSON: self.JSON_NAME,
-            self.MD5: '02c816c95f5bdb147ab3ccb6bc484a2c'
+            self.MD5: '140f972adcb7c796128970df64edfba5'
         }
         self.run_basic_test(input_dir, params, 'fusion', logging.ERROR, work_dir)
 

diff --git a/src/lib/djerba/plugins/genomic_landscape/constants.py b/src/lib/djerba/plugins/genomic_landscape/constants.py
@@ -10,6 +10,12 @@
 DONOR = 'donor'
 MSI_FILE = 'msi_file'
 CTDNA_FILE = 'ctdna_file'
+SAMPLE_TYPE = 'sample_type'
+UNKNOWN_SAMPLE_TYPE = 'Unknown sample type'
+
+# biomarker reportability
+CAN_REPORT_HRD = 'can_report_hrd'
+CAN_REPORT_MSI = 'can_report_msi'
 
 # For MSI file
 MSI_RESULTS_SUFFIX = '.recalibrated.msi.booted'

diff --git a/src/lib/djerba/plugins/genomic_landscape/genomic_landscape_template.html b/src/lib/djerba/plugins/genomic_landscape/genomic_landscape_template.html
@@ -10,20 +10,18 @@
 ${html_builder().section_cells_begin("<h2>Genomic Landscape</h2>","main")}
 
       <p >
-            Tumour Mutation Burden (TMB) was <strong>${results.get(constants.GENOMIC_LANDSCAPE_INFO).get(constants.TMB_PER_MB)}</strong> coding mutations per Mb (${html_builder().k_comma_format(results.get(constants.GENOMIC_LANDSCAPE_INFO).get(constants.TMB_TOTAL))} mutations)
-            which corresponds to the ${gl_html_builder().make_ordinal(results.get(constants.GENOMIC_LANDSCAPE_INFO).get(constants.PAN_CANCER_PERCENTILE))} 
-            percentile of the pan-cancer cohort and classified it as <strong>${results.get(constants.BIOMARKERS).get(constants.TMB).get(constants.METRIC_TEXT)}</strong>.       
-
+            Tumour Mutation Burden (TMB) was <strong>${results.get(constants.GENOMIC_LANDSCAPE_INFO).get(constants.TMB_PER_MB)}</strong> coding mutations per Mb (${html_builder().k_comma_format(results.get(constants.GENOMIC_LANDSCAPE_INFO).get(constants.TMB_TOTAL))} mutations) which corresponds to the ${html_builder().make_ordinal(results.get(constants.GENOMIC_LANDSCAPE_INFO).get(constants.PAN_CANCER_PERCENTILE))} percentile of the pan-cancer cohort and classified it as <strong>${results.get(constants.BIOMARKERS).get(constants.TMB).get(constants.METRIC_TEXT)}</strong>.
             % if results.get(constants.GENOMIC_LANDSCAPE_INFO).get(constants.CANCER_SPECIFIC_PERCENTILE) != "NA" :
-                  This TMB placed the tumour in the <strong>${gl_html_builder().make_ordinal(results.get(constants.GENOMIC_LANDSCAPE_INFO).get(constants.CANCER_SPECIFIC_PERCENTILE))}</strong> percentile of the ${results.get(constants.GENOMIC_LANDSCAPE_INFO).get(constants.CANCER_SPECIFIC_COHORT)} cohort.
+                  This TMB placed the tumour in the <strong>${html_builder().make_ordinal(results.get(constants.GENOMIC_LANDSCAPE_INFO).get(constants.CANCER_SPECIFIC_PERCENTILE))}</strong> percentile of the ${results.get(constants.GENOMIC_LANDSCAPE_INFO).get(constants.CANCER_SPECIFIC_COHORT)} cohort.
             % endif
 
-            % if results.get(constants.PURITY) > 50:
+            % if results.get(constants.CAN_REPORT_MSI):
                   The microsatellite status was <strong>${results.get(constants.BIOMARKERS).get(constants.MSI).get(constants.METRIC_TEXT)}</strong>.
             % endif 
-            This tumour had <strong>${html_builder().k_comma_format(results.get(constants.CTDNA).get(constants.CTDNA_CANDIDATES))}</strong> candidate somatic SNVs genome-wide, 
-            making the sample <strong>${results.get(constants.CTDNA).get(constants.CTDNA_ELIGIBILITY)}</strong> for OICR's plasma WGS cfDNA assay (minimum of 4,000 SNVs required). 
-             This sample shows signatures consistent with <strong>${results.get(constants.BIOMARKERS).get(constants.HRD).get(constants.METRIC_TEXT)}</strong>.
+            This tumour had <strong>${html_builder().k_comma_format(results.get(constants.CTDNA).get(constants.CTDNA_CANDIDATES))}</strong> candidate somatic SNVs genome-wide, making the sample <strong>${results.get(constants.CTDNA).get(constants.CTDNA_ELIGIBILITY)}</strong> for OICR&apos;s plasma WGS cfDNA assay (minimum of 4,000 SNVs required). 
+            % if results.get(constants.CAN_REPORT_HRD):
+            This sample shows signatures consistent with <strong>${results.get(constants.BIOMARKERS).get(constants.HRD).get(constants.METRIC_TEXT)}</strong>.
+	    % endif
       </p>
 
       <!-- other biomarkers table -->
@@ -34,12 +32,10 @@
                   <th style="width:80%">Score & Confidence</th>
             </thead>
             <tbody>
-                  % for row in gl_html_builder().biomarker_table_rows(results.get(constants.BIOMARKERS), results.get(constants.PURITY)):
+                  % for row in gl_html_builder().biomarker_table_rows(results.get(constants.BIOMARKERS), results.get(constants.CAN_REPORT_HRD), results.get(constants.CAN_REPORT_MSI)):
                   ${row}
                   % endfor 
             </tbody>
-
-
       </table>
 
 

diff --git a/src/lib/djerba/plugins/genomic_landscape/plugin.py b/src/lib/djerba/plugins/genomic_landscape/plugin.py
@@ -3,6 +3,7 @@
 """
 import csv
 import os
+import re
 
 import djerba.core.constants as core_constants
 import djerba.plugins.genomic_landscape.constants as glc
@@ -44,7 +45,8 @@ def specify_params(self):
             glc.PURITY_INPUT,
             glc.MSI_FILE,
             glc.CTDNA_FILE,
-            glc.HRDETECT_PATH
+            glc.HRDETECT_PATH,
+            glc.SAMPLE_TYPE
         ]
         for key in discovered:
             self.add_ini_discovered(key)
@@ -75,6 +77,7 @@ def configure(self, config):
         dpi = core_constants.DEFAULT_PATH_INFO
         oc = oncokb_constants.ONCOTREE_CODE
         w = self.update_wrapper_if_null(w, ipf, glc.TCGA_CODE)
+        w = self.update_wrapper_if_null(w, ipf, glc.SAMPLE_TYPE, fallback=glc.UNKNOWN_SAMPLE_TYPE)
         w = self.update_wrapper_if_null(w, ipf, oc, self.INPUT_PARAMS_ONCOTREE_CODE)
         w = self.update_wrapper_if_null(w, ppf, purple_constants.PURITY)
         w = self.update_wrapper_if_null(w, dsi, glc.TUMOUR_ID)
@@ -96,16 +99,36 @@ def extract(self, config):
         plugin_dir = os.path.dirname(os.path.realpath(__file__))
         r_script_dir = os.path.join(plugin_dir, 'Rscripts')
 
-        # Make a file where all the (actionable) biomarkers will go
+        # Make a file where all the (actionable) biomarkers will go, and initialize results
         biomarkers_path = self.make_biomarkers_maf(work_dir)
         results = tmb_processor(self.log_level, self.log_path).run(
             work_dir, data_dir, r_script_dir, tcga_code, biomarkers_path, tumour_id
         )
-        results[glc.PURITY] = wrapper.get_my_float(glc.PURITY_INPUT) * 100
+        # evaluate HRD and MSI reportability
+        purity = wrapper.get_my_float(glc.PURITY_INPUT)
+        sample_type = wrapper.get_my_string(glc.SAMPLE_TYPE)
+        sample_is_ffpe = False
+        if re.search('FFPE', sample_type.upper()):
+            sample_is_ffpe = True
+            self.logger.debug('FFPE sample detected')
+        elif sample_type == glc.UNKNOWN_SAMPLE_TYPE:
+            self.logger.warning("Unknown sample type in config; assuming non-FFPE sample")
+        else:
+            self.logger.debug('Non-FFPE sample detected')
+        if purity >= 0.5 or (purity >= 0.3 and not sample_is_ffpe):
+            results[glc.CAN_REPORT_HRD] = True
+        else:
+            results[glc.CAN_REPORT_HRD] = False
+        if purity >= 0.5:
+            results[glc.CAN_REPORT_MSI] = True
+        else:
+            results[glc.CAN_REPORT_MSI] = False
+        # evaluate biomarkers
         results[glc.CTDNA] = ctdna_processor(self.log_level, self.log_path).run(wrapper.get_my_string(glc.CTDNA_FILE))
         hrd = hrd_processor(self.log_level, self.log_path)
         results[glc.BIOMARKERS][glc.HRD] = hrd.run(
-            work_dir, wrapper.get_my_string(glc.HRDETECT_PATH)
+            work_dir,
+            wrapper.get_my_string(glc.HRDETECT_PATH)
         )
         results[glc.BIOMARKERS][glc.MSI] = msi_processor(self.log_level, self.log_path).run(
             work_dir,

diff --git a/src/lib/djerba/plugins/genomic_landscape/render.py b/src/lib/djerba/plugins/genomic_landscape/render.py
@@ -11,35 +11,27 @@ def assemble_biomarker_plot(self,biomarker,plot):
         cell = template.format(biomarker,plot)
         return(cell)
 
-    def biomarker_table_rows(self, biomarkers, purity):
+    def biomarker_table_rows(self, biomarkers, can_report_hrd, can_report_msi):
         rows = []
         for marker, info in biomarkers.items():
-            cells = [
-                hb.td(info[constants.ALT]),
-                hb.td(info[constants.METRIC_ALTERATION]),
-                hb.td(self.assemble_biomarker_plot(info[constants.ALT], info[constants.METRIC_PLOT]))
-            ]
-            if marker == "MSI" and purity < 50:
+            if marker == "HRD" and not can_report_hrd:
                 cells = [
                     hb.td(info[constants.ALT]),
                     hb.td("NA"),
-                    hb.td("Cancer cell content &#8804; 50 &#37;, below threshold to call MS score")
+                    hb.td("Cancer cell content below threshold to evaluate HRD; must be &#8805;50&#37; for FFPE samples, &#8805;30&#37; otherwise")
+                ]
+            elif marker == "MSI" and not can_report_msi:
+                cells = [
+                    hb.td(info[constants.ALT]),
+                    hb.td("NA"),
+                    hb.td("Cancer cell content below threshold to call MS score; must be &#8805;50&#37;")
+                ]
+            else:
+                cells = [
+                    hb.td(info[constants.ALT]),
+                    hb.td(info[constants.METRIC_ALTERATION]),
+                    hb.td(self.assemble_biomarker_plot(info[constants.ALT], info[constants.METRIC_PLOT]))
                 ]
             rows.append(hb.table_row(cells))
         return rows
 
-    def make_ordinal(self,n):
-        '''
-        Convert an integer into its ordinal representation::
-
-            make_ordinal(0)   => '0th'
-            make_ordinal(3)   => '3rd'
-            make_ordinal(122) => '122nd'
-            make_ordinal(213) => '213th'
-        '''
-        n = int(n)
-        if 11 <= (n % 100) <= 13:
-            suffix = 'th'
-        else:
-            suffix = ['th', 'st', 'nd', 'rd', 'th'][min(n % 10, 4)]
-        return str(n) + suffix
diff --git a/src/lib/djerba/plugins/genomic_landscape/test/genomic_landscape.ini b/src/lib/djerba/plugins/genomic_landscape/test/genomic_landscape.ini
@@ -1,4 +1,5 @@
 [core]
+author = Test Author
 
 [genomic_landscape]
 oncotree code=PAAD
@@ -7,4 +8,5 @@ purity=0.78
 tumour_id=PLACEHOLDER
 hrd_path = $DJERBA_TEST_DIR/plugins/hrd/hrdetect.signatures.json
 msi_file=$DJERBA_TEST_DIR/plugins/genomic-landscape/PLACEHOLDER.filter.deduped.realigned.recalibrated.msi.booted
-ctdna_file=$DJERBA_TEST_DIR/plugins/genomic-landscape/SNP.count.txt
+ctdna_file=$DJERBA_TEST_DIR/plugins/genomic-landscape/SNP.count.txt
+sample_type=FFPE
diff --git a/src/lib/djerba/plugins/genomic_landscape/test/plugin_test.py b/src/lib/djerba/plugins/genomic_landscape/test/plugin_test.py
@@ -60,7 +60,7 @@ def testGenomicLandscapeLowTmbStableMsi(self):
         params = {
             self.INI: self.INI_NAME,
             self.JSON: json_location,
-            self.MD5: 'b8483c476a1c17404ac81f9e3a439641'
+            self.MD5: '690242139aab0153348e7a4634d2f17c'
         }
         self.run_basic_test(input_dir, params)
 

diff --git a/src/lib/djerba/plugins/supplement/body/resources/TAR.description.versions.html b/src/lib/djerba/plugins/supplement/body/resources/TAR.description.versions.html
@@ -9,8 +9,8 @@
 Shallow whole genome sequencing (sWGS) libraries were prepared using the KAPA Hyper Prep kit with DNA extracted from FFPE, cfDNA, fresh frozen tissue (for tumour samples). 
 Paired-end sequencing was performed using the ${versions.TAR_ILLUMINA_VERSION} illumina technology to a minimum target coverage of 0.1x. 
 Alignments were performed using <a href=${versions.BWAMEM_LINK}>bwa mem</a> (${versions.BWAMEM_VERSION})
-against reference genome <a href=${versions.REFERENCE_GENOME_LINK}>${versions.REFERENCE_GENOME_VERSION}</a>. 
-and copy number amplifications are called using <a href=${versions.ICHORCNA_LINK}>ichorCNA</a>.
+against reference genome <a href=${versions.REFERENCE_GENOME_LINK}>${versions.REFERENCE_GENOME_VERSION}</a>.
+Copy number amplifications were called using <a href=${versions.ICHORCNA_LINK}>ichorCNA</a>.
 
 TAR libraries were prepared using the KAPA Hyper Prep kit with DNA extracted from FFPE, cfDNA, fresh frozen tissue (for tumour samples) or buffy coat blood specimens (for matched normal blood samples). 
 Paired-end sequencing was performed using the ${versions.TAR_ILLUMINA_VERSION} illumina technology. 

diff --git a/src/lib/djerba/plugins/supplement/body/resources/WGS.description.versions.html b/src/lib/djerba/plugins/supplement/body/resources/WGS.description.versions.html
@@ -14,16 +14,16 @@
 and annotated with <a href=${versions.VARIANTEFFECTPREDICTOR_LINK}>VariantEffectPredictor</a> (v.${versions.VARIANTEFFECTPREDICTOR_VERSION}) 
 using MANE transcripts (MANE Clinical version 1.0 when available, <a href=${versions.MANE_LINK}>MANE Select</a> version ${versions.MANE_VERSION} for all other transcripts).
 Variants were further annotated for oncogenicity and actionability by <a href=${versions.ONCOKB_LINK}>OncoKB</a>. 
-In cases where OncoKB does not use MANE Select, links in annotation use the corresponding alteration in OncoKB.
+In cases where OncoKB did not use MANE Select, links in annotation have used the corresponding alteration in OncoKB.
 Copy number variations were called using <a href=${versions.PURPLE_LINK}>Purple</a> (${versions.PURPLE_VERSION}). 
-Microsatellite (MS) Instability status is called using <a href=${versions.MICROSATELLITE_LINK}>msisensor-pro</a> (${versions.MICROSATELLITE_VERSION}) and a custom list of MS sites created by msisensor-pro for the current reference genome.
-Homologous recombination deficiency (HRD) status is called using HRDetect <a href="https://pubmed.ncbi.nlm.nih.gov/28288110/">(Davies et al. 2017)</a>, a weighted logistic regression model, using the signature.tools.lib R package <a href="https://pubmed.ncbi.nlm.nih.gov/32118208/">(Degasperi et al. 2020).</a>. HRDetect takes SNVs and in/dels from MuTect2. The proportion of deletions that are at microhomologous sites is summarized as "Microhomologous Deletions".
-The counts of SNVs are categorized into exposures based on their trinucleotide context using 
+Microsatellite (MS) Instability status was called using <a href=${versions.MICROSATELLITE_LINK}>msisensor-pro</a> (${versions.MICROSATELLITE_VERSION}) and a custom list of ${versions.MICROSATELLITE_CUSTOM_SITES} MS sites created by msisensor-pro for the current reference genome.
+Homologous recombination deficiency (HRD) status was called using HRDetect <a href="https://pubmed.ncbi.nlm.nih.gov/28288110/">(Davies et al. 2017)</a>, a weighted logistic regression model, using the signature.tools.lib R package <a href="https://pubmed.ncbi.nlm.nih.gov/32118208/">(Degasperi et al. 2020)</a>. HRDetect takes SNVs and in/dels from MuTect2. The proportion of deletions occurring at microhomologous sites has been summarized as "Microhomologous Deletions".
+The counts of SNVs were categorized into exposures based on their trinucleotide context using 
 <a href="https://cran.r-project.org/web/packages/deconstructSigs/index.html">DeconstructSigs</a> (v. 1.8.0) and SBS signatures
 as defined in <a href="https://cancer.sanger.ac.uk/signatures/downloads/">COSMIC version 1</a>. 
-HRDetect also takes in LOH and structural variants. Structural variants are first called by 
+HRDetect also takes in LOH and structural variants. Structural variants were first called by 
 <a href="https://github.com/PapenfussLab/gridss">GRIDSS</a> (v.2.13.2)
-and then passed to PURPLE (v.3.8.1) for integrated LOH calling. Structural variants are then categorized into exposures based on break-end 
+and then passed to PURPLE (v.3.8.1) for integrated LOH calling. Structural variants were then categorized into exposures based on break-end 
 characteristics using <a href="https://github.com/Nik-Zainal-Group/signature.tools.lib">signature.tools.lib</a> (v. 2.1.2) 
 and the rearrangement signature set defined in <a href="https://www.nature.com/articles/nature17676">Nik-Zainal et al. (2016)</a>. 
 </p>
diff --git a/src/lib/djerba/plugins/supplement/body/resources/WGS.disclaimer.html b/src/lib/djerba/plugins/supplement/body/resources/WGS.disclaimer.html
@@ -1,8 +1,8 @@
 <p>
 Based on a minimum tumour purity of 30%, the sensitivity for SNVs and in/dels is 96% and 89%, respectively. 
 The sensitivity for CNVs and RNA fusions is 100% and 32%, respectively. 
-The limit of detection is 10% VAF for SNVs and 20% for in/dels. The limit of detection for MSI is cellularity >50%. 
-For HRD, the sensitivity is 83% and the specificity is 90%. The lower limit of detection is >50% cellularity in FFPE samples and >30% cellularity in fresh frozen samples.
+The limit of detection is 10% VAF for SNVs and 20% for in/dels. The limit of detection for MSI is cellularity &ge;50%. 
+For HRD, the sensitivity is 83% and the specificity is 90%. The lower limit of detection is &ge;50% cellularity in FFPE samples and &ge;30% cellularity in fresh frozen samples.
 For LOH, the sensitivity and specificity are both 100%. LOH is currently reported for autosomes; LOH on the X chromosome is not reported.
 Although whole genome sequencing encompasses all genes in a specimen, this report is restricted to cancer genes defined by OncoKB as of the date the report is issued.
 This test was developed and its performance characteristics determined by OICR Genomics. It has not been cleared or approved by the US Food and Drug Administration.