PM-OICR TGL

Collaborate. Translate. Change lives.

This document is a supplement to our website and lab protocols. Lab and informatic protocols are summarized for easy insertion into grant applications, proposals and clinical protocols for ethics review committees (CAPCR, iREB). Full lab and informatic protocols and workflow diagrams are available at https://labs.oicr.on.ca/translational-genomics-laboratory/

PM-OICR TGL Manual for Grant Applications, Proposals and Clinical Research Protocols

PM-OICR TGL

Page 1 of 28 Version 1.9

Contents 1. PM-OICR TGL Mandate ......................................................................................................................... 3

1.1 Structure of OICR Genomics ......................................................................................................... 4

Figure 1.1: OICR Genomics Departmental Structure. ........................................................................... 5

2. Recommended Core Clinical Data Elements ......................................................................................... 5

3. Shipment and/ or Sample Receipt at PM-OICR TGL ............................................................................. 6

(a) Personal Health Information (PHI) ............................................................................................ 7

4. Pathology/Correlative Lab Requisition Guidance ................................................................................. 7

4.1 Archival Tissue Recommendations ............................................................................................... 8

(a) Tissue and Biopsy Cores (Fresh frozen and FFPE) ..................................................................... 8

4.2 Blood/Plasma Collection Recommendations ................................................................................ 9

5. Lab Protocols ......................................................................................................................................... 9

5.1 Fresh Frozen Tissue Embedding in OCT ........................................................................................ 9

5.2 Plasma and Buffy Coat Separation from Whole Blood Protocol ................................................ 10

5.3 DNA/RNA Isolation from Buffy Coat Protocol ............................................................................. 10

5.4 Circulating Tumor DNA (ctDNA) Isolation from Plasma Protocol ............................................... 11

5.5 DNA/RNA Co-isolation from FFPE Slides Protocol ...................................................................... 11

5.6 CAP RNA Sequencing Lab Protocol ............................................................................................. 11

5.7 Exome Sequencing Lab Protocol ................................................................................................. 11

5.8 Shallow and/or CAP Whole Genome Sequencing (WGS) Lab Protocol ...................................... 12

5.9 Methylation EPIC Exome Sequencing Lab Protocol .................................................................... 12

5.10 Methylation Array Lab Protocol .................................................................................................. 12

5.11 cfMeDIP Cell Free Methylated DNA Immunoprecipitation (IP), Targeted Sequencing Panels and Shallow Whole Genome Lab Protocol ................................................................................................... 13

5.12 ctDNA Targeted Sequencing Panels Lab Protocol ....................................................................... 13

(a) ctDNA Target Panel Validation Recommendations .............................................................. 14

6. Informatics .......................................................................................................................................... 14

6.1 Pre-analytic Library Validation Informatics ................................................................................. 14

6.2 Exome Informatics ...................................................................................................................... 14

Figure 6.2a: Tumor only exome pipeline. ........................................................................................... 16

Figure 6.2b: Tumor- normal (paired) exome pipeline. ....................................................................... 16

6.3 Low Pass (Shallow) Whole Genome and CAP Whole Genome Informatics ................................ 17

PM-OICR TGL

Page 2 of 28 Version 1.9

Figure 6.3a: Shallow whole genome workflow. .................................................................................. 17

Figure 6.3b: CAP Tumor-normal paired whole genome workflow. .................................................... 18

6.4 RNA Informatics .......................................................................................................................... 18

Figure 6.4a: Tumor transcriptome pipeline. ....................................................................................... 19

Figure 6.4b: CAP Tumor transcriptome pipeline. ............................................................................... 19

6.5 Methylation Sequencing Informatics .......................................................................................... 19

Figure 6.5a: Tumor methylation pipeline. .......................................................................................... 20

6.6 cfMeDIP Informatics ................................................................................................................... 20

(a) Figure 6.6a: cfMeDIPs workflow. ............................................................................................ 21

6.7 ctDNA Informatics ....................................................................................................................... 21

6.8 Infinium Methylation EPIC Array Informatics ............................................................................. 21

7. Contracts: Material and Data Transfer Agreements, Statement of Work Forms ............................... 22

7.1 Collaborations between UHN & PM-OICR TGL, Statement of Work Forms ............................... 22

7.2 Collaborations between PM-OICR TGL and an external institute ............................................... 22

8. Ethics ................................................................................................................................................... 23

8.1 UHN CAPCR Approved Protocols ................................................................................................ 23

8.2 External Ethics Board Approved Protocols (Non-CAPCR) ........................................................... 23

9. Publication Policy ................................................................................................................................ 24

9.1 Authorship Policy ........................................................................................................................ 24

(a) Suggested Authorship Listing for Investigator Initiated Trial (Assuming 20 authors): ........... 24

9.2 Publication Acknowledgements .................................................................................................. 25

10. Revision History .............................................................................................................................. 25

11. References: ..................................................................................................................................... 26

PM-OICR TGL

Page 3 of 28 Version 1.9

1. PM-OICR TGL Mandate

The Princess Margaret Cancer Centre-Ontario Institute for Cancer Research Translational Genomics Lab (PM-OICR TGL) is a joint collaborative workspace located in the MaRS complex at University Avenue and College Street in downtown Toronto. PM-OICR TGL is a department of OICR Genomics. OICR genomics includes PM-OICR TGL, OICR Genome Research Platform (GRP) and the Princess Margaret Genomics Core. TGL enhances rapid access to genomics technologies, facilitating genomic interpretation and reporting in clinical oncology. TGL provides research support to the Princess Margaret Cancer Centre (UHN) and affiliated cancer research institutions as part of OICR’s Adaptive Oncology Program. TGL is transitioning to a College of American Pathologists (CAP) certified laboratory integrating College of Medical Laboratory Technologists of Ontario (CMLTO) licensed laboratory professionals. CAP certified whole genome and transcriptome assays are anticipated to be available in the 1st quarter of 2020. TGL is formulated to accelerate clinical oncology research through partnerships and drive key initiatives:

• Enable genome-wide multi-omic assays from pathology specimens including Formalin Fixed Paraffin Embedded (FFPE) tumors, fresh frozen tissues, blood and plasma (liquid biopsy);

• Identify molecular patterns associated with patient outcome and clinical variables from standard of care and second line therapies;

• Share and continuously improve integrated informatics methods to facilitate robust tumor phenotypic classification;

• Facilitate actionable mutation detection through an emphasis on multiplatform molecular diagnostics, including methylation, transcriptome, exome, plasma targeted sequencing and cell free methylated DNA Immunopreciptiation (cfMeDIP) and whole/shallow genome sequencing;

• Facilitate rapid dissemination of patient tumor genomic reports and annotation, integrating pathology, imaging (H&E and CT scans) and genomic interpretation into a unified database hosted at cbioportal.ca

PM-OICR TGL

Page 4 of 28 Version 1.9

1.1 Structure of OICR Genomics

The Princess Margaret Cancer Centre (PM) and OICR have formed a joint Genomics Program with the mission to support the continuum of basic, translational, and clinical research. The Genomics Program includes 1) the Princess Margaret Genomics Centre as an incubator for new technologies such as epigenetics and single cell transcriptomics (https://pmgenomics.ca), 2) the OICR Genome Research Platform (GRP) for scaling up and productionizing mature technologies such as whole genome sequencing and RNA-seq (https://genomics.oicr.on.ca), and 3) the Translational Genomics Laboratory that is undergoing accreditation by the College of American Pathologists and the International Standards Organization, specifically ISO15189, required for clinical reporting of genomic assays (https:// https://labs.oicr.on.ca/translational-genomics-laboratory). To coordinate these three data generation arms, an overarching Quality Assurance and Program Management department ensures adherence to accredited standards across the program including implementation of quality control standards and data systems necessary to deliver high-quality, consistent data. The Genome Sequence Informatics (GSI) department implements standardized pipelines and data systems, as technologies mature from basic through translational to clinical-grade assays. Together, the Genomics Program brings together >50 highly trained personnel with expertise necessary to deliver the myriad assays required by our program. OICR genomics infrastructure includes 12 Illumina next-generation sequencing devices (2 NovaSeq 6000s, 3 HiSeq 2500s, 4 NextSeq 550s, 3 MiSeqs) representing capacity equivalent to >9,000 30X whole genome sequences per year. We also maintain 2 nanopore-based DNA sequencers (Oxford Nanopore PromethION and MinION), 4 laboratory robotics systems, 2 10X Genomics Chromium Devices, 1 Nanostring nCounter device, and ancillary equipment necessary to carry out conventional genomics assays. Associated, professionally-managed high-performance computing clusters include the OICR Informatics platform and UHN HPC4Health cluster that together comprise >15,000 CPU cores and >10 Pb of storage. Our organizational structure is detailed below:

PM-OICR TGL

Page 5 of 28 Version 1.9

Figure 1.1: OICR Genomics Departmental Structure.

2. Recommended Core Clinical Data Elements

PM-OICR TGL recommends the capture of core data elements endorsed by the Centre for Medical Technology Policy and Molecular Evidence Development Consortium [1]. When possible, it is advised to collect the following data elements:

• Demographics: gender, ethnicity, race, cause of death • Medical history: prior malignancies • Physical exam at first diagnosis: height, weight, performance status • First diagnosis of cancer of interest: basis of diagnosis, cancer site and histology, stage

and grade, site and type of tissue sampling, prognostic biomarkers (presence/absence and levels), additional molecular diagnostic testing and performing laboratory

• Treatment episode: therapeutic agent and/or modality, intent of treatment, reason for ending treatment

• Outcomes (for each assessment episode): disease response, method of response evaluation, sites of any progression/recurrence, vital status, performance status and weight

• Dates: year of birth, date of death, date of diagnosis of any prior malignancies, date of physical evaluation at diagnosis, date of definitive diagnosis, beginning and ending dates

PM-OICR TGL

Page 6 of 28 Version 1.9

of treatment (for each treatment), date of assessment of outcome (for each assessment)

We are supportive of clinical data capture through Labkey. Upon request, TGL will provide tables for curated collection of clinical information from Medidata.

We recommend Minimal Common Oncology Data Elements (mCODE) for Electronic Health Records (EHR) to standardize capture of clinical data elements. Use of mCODE enables enhanced computational analysis of clinical variables across large datasets.

3. Shipment and/ or Sample Receipt at PM-OICR TGL

Materials may be transferred to PM-OICR TGL from UHN’s Correlative Studies Program or UHN BioBank from the following addresses, or other sites as designated in study protocol: Correlative Studies Program c/o Vanessa Spears Princess Margaret Cancer Centre 610 University Avenue, Suite 7-409 Toronto, Ontario M5G0A3 416.946.4501 X2562 Vanessa.speers@uhn.ca UHN BioBank Toronto General Hospital, Eaton Wing 11th floor, RM1126 200 Elisabeth Street Toronto, Ontario M5G2C4 416.340.4800 X4744 Biospecimen.sciences.program@uhn.ca To PM-OICR TGL in person, or via scheduled pickup (research only samples): PM-OICR TGL c/o Dax Torti Ontario Institute for Cancer Research 661 University Avenue, Suite 510 Toronto, Ontario

PM-OICR TGL

Page 7 of 28 Version 1.9

M5G0A3 Dax.torti@oicr.on.ca; tglsamples@lists.oicr.on.ca 647.260.7938 To Tissue Portal (for CAP assay stream sample submissions only): Attn: Ilinca Lungu Tissue Portal Sample Receiving Department of Diagnostic Development Ontario Institute for Cancer Research 661 University Avenue MaRS Centre 6th Floor Suite 6-46 Toronto, Ontario M5G 0A3 Perishable materials including fresh frozen tissues, blood and derivatives (plasma, peripheral blood mononuclear cells (PBMCs), circulating free DNA (cfDNA) and circulating tumor DNA (ctDNA) must be shipped/packed on dry ice. Formalin fixed paraffin embedded (FFPE) tissues, blocks and/or sections may be shipped/packed in slide boxes at room temperature for transport. All materials must contain de-identified study codes only and be accompanied with a sample manifest (sample submission sheet) available online at https://labs.oicr.on.ca/translational-genomics-laboratory/. Completed sample submission sheets must be emailed to dax.torti@oicr.on.ca and tglsamples@lists.oicr.on.ca prior to sample receipt.

(a) Personal Health Information (PHI)

All materials and data sent to PM-OICR TGL must be de-identified, removing any reference to direct identifiers including patient name, medical record number (MRN), chart number, and/or surgical numbers. Delete and/or redact all information from pathology reports or digital files prior to transfer. De-identified biobank codes and study identification codes are acceptable. Please maintain a minimum of two identifiers on any tissue. OICR’s policies on privacy (PR-INS.101.001) and privacy breach (PR-INS.301) are available upon request from program managers, or through OICR’s Privacy Officer. 4. Pathology/Correlative Lab Requisition Guidance

TGL only accepts tissues, buffy coat and/or plasma. If you must extract nucleic acids, please follow our recommended protocols. For all correlative studies, always collect a germline normal reference or buffy coat along with tumor tissue, for ctDNA profiling, only 1 buffy coat (1mL aliquot) is required. A germline normal control will ensure high confidence somatic calls.

PM-OICR TGL

Page 8 of 28 Version 1.9

If integrating TCR receptor profiling (academic collaborations only), serial buffy coat time points are required.

4.1 Archival Tissue Recommendations

PM-OICR TGL recommends the following parameters for tissue requests: 10 micron sections, minimum n=10, 1 hematoxylin and eosin (H&E) at the top of the sectioning stack to establish tumor cellularity, 10 unstained slides cut onto uncharged slides, air dried (if study includes immunohistochemistry assays, charged slides are required), and 1 H&E at the end of the sectioning stack to verify tumor cellularity. We recommend 10 micron sections, as nuclei on average, are 6 microns in diameter and may be damaged when tissues are cut at <10 microns. Please notify the consulting pathologist to circle tumor regions on H&E slides and estimate % tumor cells (cellularity) within tumor area and % necrotic cells within the tumor area. Circled tumor regions should have a minimum cellularity of 40%. Please provide a de-identified pathology report containing tissue site, diagnosis, viable tumor (%) and % necrosis for the whole section and the subsection for analysis. This information will be integrated into shallow whole genome assessments of tumor purity and ploidy. Where possible, TGL will use OncoTree codes for tissue classification. If H&E images are scanned at your institute, we request a minimum of 20X resolution (TGL scans at 40X) including pathology markup. Please forward image files as “.SVS” format (Aperio Image File Format, type code 3 recommended) via secure file transfer; images will be incorporated into cBioPortal.ca (Aperio eSlide manager interface). Images may be stored in Microsoft Azure Blob storage inside a UHN virtual network on Azure. All data (including metadata) written to Azure Storage is encrypted using Storage Service Encryption. Project access permissions are controlled through Keycloak user authentication at the project level and administered through TGL’s project manager. TGL prefers to receive marked H&E, but will accept digital scans with tumor markup to guide tissue extraction. If the entire tissue section is tumor, the entire tissue may be circled and scored. Ideally a total surface area of >150 mm2 (15mm2 tumor surface area X 10 slides) will yield sufficient DNA and RNA for exome and RNA sequencing libraries; macro dissection of multiple slides may be required. A maximum of 600mm2 of tumor tissue (100mm2 tumor surface area X 6 slides) may be extracted over 1 purification column set. This information may be detailed in PM-OICR TGL submission sheets. De-identified study codes must be used on all documentation, we prefer a minimum of two unique identifiers on all samples received. PM-OICR TGL requests de-identified pathology reports be forwarded through secure file transfer.

(a) Tissue and Biopsy Cores (Fresh frozen and FFPE)

We recommend 2-3 cores, ideally 14G (2.1mm), but as small as 19G (1.1mm). FFPE cores may be sectioned as specified in section 4.1 and must have pathology review. Fresh frozen cores should be embedded in OCT medium (see section 5.1) prior to long term storage at -80C; freezer burn may damage unembedded tissues. Please provide a de-identified pathology report. For studies requiring CAP certified assays and clinical reports, tissue blocks and H&E’s will be received by Tissue Portal at OICR. Contact dax.torti@oicr.on.ca for further instruction.

PM-OICR TGL

Page 9 of 28 Version 1.9

4.2 Blood/Plasma Collection Recommendations

Streck Cell Free DNA BCT blood collection tubes (10ml) PN#218961 (Research Use Only, RUO), or PN#218996 (CE, not available in USA) are recommended for genomic applications including peripheral blood mononuclear cell (PBMC) isolation for germline (normal) exome, and cell free DNA assays (cfDNA) including targeted ctDNA panels and cfMeDIP assays. 3-4X10 ml collection tubes should be drawn at each time point in a protocol and processed into plasma and PBMC fractions ideally within 30 minutes, but up to 2 hours after collection. Blood collected in Streck tubes is stable at room temperature, but storage at 4°C is preferable. cfDNA isolated from blood collected in Streck tubes is more resilient to degradation from sub-optimal storage conditions and/or delays in plasma isolation. Please see Risberg et al [2] for a full review of ctDNA stability collected in EDTA versus Streck collection vials. Alternatively, K3EDTA (BD366450, or other vendor) blood collection tubes (10ml) may be used. 3-4X10ml collection tubes should be drawn at each time point in a protocol. K3EDTA tubes are susceptible to leukocyte lysis which may diminish detection of mutant allele fractions in ctDNA and cfMeDIP applications. Blood collected in K3EDTA tubes should preferably be stored at 4°C, and processed to plasma and PBMC fractions within 30 minutes or a maximum of 2 hours after collection. FDA guidelines specify that the total amount of blood drawn over an 8-week period may not exceed the lesser of 50mL or 3mL/kg, or be collected more than 2 times per week. When possible, blood should be drawn at the same time as routine blood tests. 5. Lab Protocols

All TGL protocols are available for distribution and are not copyright, with the exception of cell free methylated DNA Immunoprecipitation (cfMeDIP) assays. Protocols may be modified and/or distributed to collaborators. If you require a protocol and can’t locate it on our website, contact dax.torti@oicr.on.ca. The cfMeDIP protocol is licensed assay from UHN, licensing restrictions may apply.

5.1 Fresh Frozen Tissue Embedding in OCT

Prior to embedding tissues in OCT, chill the following items on dry ice for 1 hour: cryostat heat extractor rubber press, cool plate and forceps. OCT compound is applied to a room temperature plastic mount. The tissue is placed into the centre of the OCT filled plastic mount and additional OCT is added on top of tissue. The plastic mount/tissue is immediately transferred to the cool plate on dry ice and the heat extractor placed on top of the OCT embedded sample. The rubber press is pushed down on the heat extractor to remove air bubbles for 1 minute. Complete freezing should occur after 3 minutes. The embedded sample

PM-OICR TGL

Page 10 of 28 Version 1.9

is removed from the plastic mount, and excess OCT is sliced off with a razor blade. The embedded tissue is wrapped in parafilm, then aluminum foil, placed in a Ziplock bag and stored at -80°C for a minimum of 1 hour before sectioning. To prevent ice crystal formation, apply a few drops of OCT to any cut surfaces and freeze to seal tissues; failure to do so may destroy/prevent future histological review and/or use of tissue for genomics applications.

Fresh frozen tissues are initially sectionedand stained with H&E for review by a pathologist. These processes are performed within OICR Tissue Portal as part of OICR Genomic’s CAP Assay Application Workflow. Serial sections are immediately macrodissected and lysed in Qiagen RLT Plus buffer + β-mercaptoethanol and stored at -80° C prior to extraction. H&E images may be digitally reviewed through cBioPortal.ca using Aperio Image analysis tools.

5.2 Plasma and Buffy Coat Separation from Whole Blood Protocol

Up to four 10 ml tubes of whole blood collected in STRECK Cell-Free DNA BCT (preferred), K3EDTA or ROCHE cell free DNA collection tubes may be processed for buffy coat and plasma fractions within 30 minutes or up to 2 hours post collection, keep samples at 4°C prior to processing. After thorough inversion of the blood collection tube, whole blood is transferred to a new 15ml conical tube and spun at 1900g at 4°C for 10 minutes. The plasma layer is transferred to a new 15ml conical tube. Plasma layers from two whole blood collection tubes may be combined into one conical tube. After collection of the plasma layer, the buffy coat layer is pipetted into 1.5ml tubes without disturbing the erythrocyte layer. The plasma collected during the first centrifugation is spun again at 16,000g for 10 minutes. Without disturbing the pellet of cellular debris, the plasma is transferred to a new 15ml conical tube, leaving 0.5ml behind. The purified buffy coat and plasma are immediately stored at -80°C or in liquid phase nitrogen for long term storage. PM-OICR TGL prefers isolated plasma in 15 ml conical tubes, but will accept 1.5 ml aliquots for processing; ideally 10 mls of plasma are required for cfDNA purification. Recommended protocols are available on our website. For CAP Assay streams, plasma will be extracted within Tissue Portal at OICR.

5.3 DNA/RNA Isolation from Buffy Coat Protocol

DNA and RNA is co-isolated from 150-250 ul of buffy coat (1ml vial) using the Qiagen AllPrep DNA/RNA/miRNA Universal Kit according to manufacturer’s directions.

For CAP assay streams, buffy coat will be extracted within Tissue Portal at OICR. Tissue portal buffy coat extractions utilize the Qiagen Gentra Purgene protocol according to manufacturer’s directions; only DNA is isolated.

PM-OICR TGL

Page 11 of 28 Version 1.9

5.4 Circulating Tumor DNA (ctDNA) Isolation from Plasma Protocol

cfDNA is isolated using the Qiagen QIAamp Circulating Nucleic Acid Kit according to manufacturer’s protocol. Briefly, plasma is spun at 16,000g to remove residual cellular debris. Plasma is treated with proteinase K and digested at 60°C for 1 hour. Lysate is processed through QIAamp mini column using a vacuum manifold, and washed successively with kit wash buffers, prior to DNA elution. Genomic contamination of isolated cfDNA is assessed using Agilent TapeStation genomic tapes and/or cell free DNA tapes.

5.5 DNA/RNA Co-isolation from FFPE Slides Protocol

DNA and RNA is co-isolated from 150mm2-600mm2 of macro dissected tumor surface area from 10 micron sections. Macro dissected material is deparaffinized using CitriSolv reagent, proteinase K digested, and DNA pellets and RNA supernatant purified over Qiagen AllPrep FFPE DNA and RNA kit columns. Isolated DNA is RNase treated, and RNA is DNase treated. Isolated material is suitable for all sequencing protocols. Full protocol details and modifications for FFPE RNA and DNA co-isolation are available on our website.

For CAP assay streams, FFPE tissue will be macro dissected and extracted within Tissue Portal at OICR.

5.6 CAP RNA Sequencing Lab Protocol

RNA libraries are synthesized from 200 ng of Total RNA using the Illumina TruSeq Stranded Total RNA with Ribozero Gold Sample Prep kit. Total RNA is depleted of ribosomal RNA (including mitochondrial ribosomal RNA), first and second strand cDNA is synthesized, A-tailed, adapter ligated, and PCR amplified. Full protocol details and modifications for FFPE RNA are available on our website. Tumor RNA is sequenced on the Illumina NextSeq550 platform, V2 chemistry and reagents, to read depth of 80 million clusters, 160 million paired end reads, 75bp X 75bp. Sequencer selection may vary depending on project specifications. All libraries are validated on the MiSeq platform prior to deep sequencing.

5.7 Exome Sequencing Lab Protocol

TGL is no longer accepting new exome sequencing projects. Existing projects may continue to submit exomes to ensure analysis and cohort integrity. Please contact PMGC or GRP departments for exome projects.

Tumor-normal exome library pairs are generated from 100 ng of DNA from fresh frozen or FFPE tumor material and/or normal buffy coat DNA. Pre-capture libraries are synthesized using a modified KAPA Hyper Prep Kit protocol prior to capture using a modified Agilent XT V6 +

PM-OICR TGL

Page 12 of 28 Version 1.9

COSMIC exome workflow. Briefly, DNA is sheared, end repaired, A-tailed, adapter ligated and PCR amplificatied. Exome probes are hybridized to DNA libraries for 16 hours and washed, prior to on-bead amplification and cleanup. Full protocol details are available on our website. Matched normal DNA is sequenced to a depth of 50X coverage, Tumor DNA, 250X coverage on Illumina NovaSeq, HiSeq2500 or NextSeq550 platforms, V1, V4, and V2 chemistry and reagents respectively. Sequencer selection may vary depending on project specifications. All libraries are validated on the MiSeq platform prior to deep sequencing.

5.8 Shallow and/or CAP Whole Genome Sequencing (WGS) Lab Protocol

Whole genome libraries are prepared using a modified Kapa HyperPrep synthesis protocol and IDT dual index adapters. Briefly, 100 ng of DNA is sheared, prior to end repair, A-tailing, adapter ligation, and PCR amplification. Shallow WGS (sWGS) Libraries are sequenced to a depth of 0.001X on the NextSeq550 platform, V2 chemistry and reagents, 75bpX75bp or on Illumina MiSeq platform, V2 chemistry and reagents, 150bpX150bp. Whole Genome libraries (WGS) are sequenced on the NOVA seq platform, 150bpX150bp, V1 chemistry and reagents to a mean depth of 80X tumor, 30X normal. All WGS libraries are validated on the MiSeq platform prior to deep sequencing. The launch of clinical reporting is anticipated in the first quarter of 2020.

5.9 Methylation EPIC Exome Sequencing Lab Protocol

This protocol is no longer offered.

TruSeq Methyl Capture EPIC exomes are prepared from 500ng of tumor DNA according to manufacturer’s directions. DNA is fragmented, end repaired, A-tailed, and adaptors ligated. Methylation EPIC probes are hybridized overnight, prior to bisulfite conversion, PCR amplification and cleanup. Libraries may be sequenced on Illumina NovaSeq, HiSeq2500 or NextSeq550 platforms, V1, V4, and V2 chemistry and reagents respectively. Sequencer selection may vary depending on project specifications. All libraries are validated on the MiSeq platform prior to deep sequencing.

5.10 Methylation Array Lab Protocol

Illumina Infinium Methylation EPIC BeadChip arrays are processed following manufacturer’s instructions. 250 ng of DNA (from fresh frozen or FFPE) is bisulfite converted using EZ DNA Methylation kit (Zymo research), FFPE DNA is repaired (Infinium HD FFPE Restore, if applicable), followed by whole genome isothermal amplification, enzymatic fragmentation, DNA precipitation and resuspension. Prepared samples are loaded on beadchips and hybridized for 16-20 hours at 48°C. Hybridized samples are washed prior to single-base extension with labelled fluorophores, followed by washing and scanning on the Illumina NextSeq550 system.

PM-OICR TGL

Page 13 of 28 Version 1.9

5.11 cfMeDIP Cell Free Methylated DNA Immunoprecipitation (IP), Targeted Sequencing Panels and Shallow Whole Genome Lab Protocol

Synthesis of cfMeDIP and targeted ctDNA libraries may be created from a single library preparation and require a minimum of 50ng of cfDNA; cfMeDIP-only library stream requires a minimum of 10 ng of cfDNA. Pre-capture libraries are synthesized using a modified KAPA Hyper Prep Kit protocol. cfDNA and Arabidopsis internal control DNA is end repaired, A-tailed and adapter ligated (unique molecular index, UMI). In dual assay mode, the pre-capture unamplified library is split into 5-methylcytosine IP fraction and an input control or pre-capture library for targeted ctDNA panel capture. 5-methylcytosine antibody (Diagenode Mag MeDIP kit) is used to selectively enrich cfDNA by immunoprecipitation of the IP sample via a modified manufacturer’s protocol, prior to library amplification and sequencing. Shallow whole genome and targeted ctDNA assays may be conducted on pre-capture libraries following methods detailed in section 5.12. Libraries may be sequenced on Illumina NovaSeq, or NextSeq550 platforms, V1, V4, and V2 Chemistry and reagents respectively, to read depth of 60 million clusters, or 120 million paired end reads. Sequencer selection may vary depending on project specifications. All libraries are validated on the MiSeq platform prior to deep sequencing.

5.12 ctDNA Targeted Sequencing Panels Lab Protocol

ctDNA libraries are prepared from a minimum of 50 ng, but ideally 100ng of circulating free (cf) DNA isolated from plasma. Higher inputs are recommended to increase the likelihood of detecting tumor ctDNA, please see Kis et al for further details [3]. It is strongly recommended to include a buffy coat control for each patient’s serial biopsy set. Sheared genomic DNA from buffy coat will be used for germline variant filtering. Pre-capture libraries are synthesized using a modified protocol based on the KAPA Hyper Prep Kit, prior to capture using Integrated DNA Technology (IDT) gene probe sets. Briefly, DNA is end repaired, A-tailed, adapter ligated (unique molecular index, UMI), and PCR amplified. Pre-capture libraries will be sequenced for copy number variation and/or interpretation of purity/ploidy. Gene specific probes are hybridized to DNA libraries for 4 hours and washed, prior to on-bead amplification and cleanup, following IDT’s Hybridization capture of DNA libraries using xGEN Lockdown Probes and Reagents protocol, version 2. Captured ctDNA is sequenced to a depth of 10,000-20,000X and may be sequenced on Illumina NovaSeq, HiSeq2500 or NextSeq550 platforms, V1, V4, and V2 Chemistry and reagents respectively. Optionally, genomic DNA or germline normal DNA isolated from PBMCs may be used as a normal reference library for sWGS workflows and/or to enhance variant filtering of targeted ctDNA panels; these pre-capture and post-capture libraries may be sequenced at low depth. Sequencer selection may vary depending on project specifications. Custom target panels may require validation to determine sensitivity and specificity.

PM-OICR TGL

Page 14 of 28 Version 1.9

(a) ctDNA Target Panel Validation Recommendations

cfDNA (ng yield) may be limiting within a study cohort, and vary by tumor type and stage of disease. The available yield from plasma may pre-determine the limit of detection (LOD) of the assay. To ensure abundant cfDNA for assay, it is recommended to collect a minimum of 3 Streck tubes per time point. The detection sensitivity for a given mutation is limited to 0.2% (fraction of reads) due to sequencer error rates inherent to Illumina Technology. Sequencing errors are present in data at 0.1% of reads; these errors may be corrected by error suppression algorithms, applied as detailed in section 6.7. To establish a 5% LOD, a ctDNA library must be synthesized from a minimum of 10ng (preferably 40-100ng) of cfDNA and sequenced to 10,000X raw sequence depth, resulting in 5,000X de-duplicated depth. For 1% LOD, a minimum of 40-100ng of cfDNA is required for library synthesis. 1% LOD requires 20,000X raw sequence depth to achieve 10,000X de-duplicated depth.

All panels must be validated for LOD. Optimal validations include one or more cfDNA or tumor genomic DNA samples with known VAF frequencies and matched buffy coat genomic DNA. Collaborators ideally will source representative samples from biobanks or existing study cohorts. Each validation cfDNA and/or tumor DNA sample will be serially diluted in respective reference buffy coat DNA for a total of 4 serial dilutions for 5% LOD, or 6 serial dilutions for 1% LOD to determine true assay and sample cohort LOD. CAP validation requests may require 20+ reference samples for validation of a new target panel, preferably with existing CAP/OLA certified results for reference comparison.

6. Informatics

6.1 Pre-analytic Library Validation Informatics

All sequencing libraries including sWGS, WGS, exome, ctDNA, cfMeDIP and total RNA libraries are validated for quality prior to deep sequencing. PM-OICR TGL performs pre-analytic sequencing on the MiSeq platform sequencing each library to a minimum read depth of 10,000 clusters, 150bpX8bpX8bpX150bp. Total RNA libraries are evaluated for ribosomal contaminant levels using RSeQC v2.6.4 [4]. Transcriptome libraries from FFPE or fresh frozen samples are failed prior to deep sequencing if (a) ribosomal sequences are in excess of 35% of reads, (b) reads per start point exceed 2.0 and (c) if % coding reads are <5%. Exome libraries are validated for insert size and duplicate levels using FastQC [5]. Libraries with insert sizes <150bp are failed, or queued for repeat library synthesis. cfMeDIP IP libraries are assessed for CpG relH and CpG GoGe enrichment, AT dropout, and methylated Arabdopsis enrichment relative to control library. Additional quality control metrics generated by FastQC are captured in our quality control database, Shiny TGLQC, for review of sequencing library quality by lab technicians.

6.2 Exome Informatics

PM-OICR TGL

Page 15 of 28 Version 1.9

Sequence reads are analyzed with FastQC [5] and aligned against human genome reference build GrCh37 (hg19) using BwaMem v 0.7.12 to generate raw sequence alignments in BAM format [6]. Preprocessing, which includes PCR-duplicate marking, indel re-alignment and base quality recalibration is performed using Picard v1.72 and GATK v3.6.0 [7]. Quality control metrics are captured within our quality control database, Shiny TGLQC. When available, preprocessing will be performed in matched tumor/normal pairs to improve indel re-alignment (see figure 6.2b). Haplotype Caller [8], MuTect1 v1.1.7 [9] and Strelka v1.0.13 [10] will be run to create raw variant call files (VCFs). In the absence of matched blood (buffy coat)/normal tissue, a pooled reference is constructed from aggregate blood (buffy coat) samples and used to normalize depth of coverage in tumor samples for CNV analysis using CNVKIT v0.9.1 [11] and MuTect2.

Raw VCF files will be annotated with Variant Effect Predictor v92 [12] which utilizes ensemble gene definitions. Germline and somatic variants will be annotated with GnomAD r2.0.1 allele frequencies [13] a database of variant frequencies in a healthy population, in order to remove common variants. Variants are filtered against GnomAD <0.001 (0.1%)[13], VAF>10% and a TGL frequency database of variants (<10%). Variants will also be annotated against known cancer hotspots v2 (CancerHotspots.org) both at the variant level and gene level [13, 14]. Analysis will include actionable /oncogenic driver analysis using the Precision Oncology Knowledge Base (OncoKB) and pathogenic database ClinVar [15, 16] . Global mutation signatures defined in Stratton et al. [17] will also be assessed using deconstructSigs v1.8.0 [18]. Through this analysis, actionable variants will be classified according to standard therapeutic intervention, investigational therapeutic implication, hypothetical therapeutic intervention, and standard therapeutic implications (resistance) including oncodriver annotations of inconclusive, likely neutral, likely oncogenic, oncogenic and unknown. Additional analysis may be applied to detect allele specific copy number profiles, loss of heterozygosity, and to estimate ploidy/cellularity using Sequenza for matched tumor/normal pairs [19]. Mutation burden will be calculated as the number of non-synonymous mutations per callable megabase; MuTect v1.1.7 [9] wig coverage file is used to determine callability. Clinical protocols that allow for data sharing as defined in patient consent forms, or as approved by ethical review boards may use a secured instance of cBioPortal [20, 21] for visualization of genomic data cohorts.

PM-OICR TGL

Page 16 of 28 Version 1.9

Figure 6.2a: Tumor only exome pipeline.

Figure 6.2b: Tumor- normal (paired) exome pipeline.

PM-OICR TGL

Page 17 of 28 Version 1.9

6.3 Low Pass (Shallow) Whole Genome and CAP Whole Genome Informatics

Reads are aligned using BwaMem v0.7.9a to generate raw sequence alignments in BAM format [6]. Aligned sequences are pre-processed using GATK v3.5 and Picard v1.9.1; pre-processing includes PCR duplicate marking, indel re-alignment in matched tumour-normal pairs and base recalibration [7]. Picard CollectWgsMetrics and CollectInsertSizeMetrics tools are used for quality control. Depth of coverage is estimated using Bedtools v2.23.0 [22]. Copy number profiling is performed using QDNAseq v1.14.0 using a bin size of 50 kb [23]. Loss of heterozygosity is estimated by comparing the heterozygous single-nucleotide polymorphism distribution profiles generated from Mutect v1.1.5 [9] against a reference SNP distribution. Tumor content/fraction and large scale copy number alterations are predicted using ichorCNA [24].

CAP whole genomes require tumor (80X) and normal (30X) sequencing depth. Mutation burden, somatic mutations, mutation signatures, tumor cell purity and ploidy, loss of heterozygosity, somatic copy number variants, and structural variants will be reported. All analysis will be conducted on genome build hg38.p12.

Figure 6.3a: Shallow whole genome workflow.

PM-OICR TGL

Page 18 of 28 Version 1.9

Figure 6.3b: CAP Tumor-normal paired whole genome workflow.

6.4 RNA Informatics

PM-OICR TGL will process Total RNA through FastQC [5], STAR aligner v2.6.0c [25], ReSeQC v2.6.4 [5] followed by RNA abundance quantification using RSEM v1.3.0 [26] to generate an expression matrix used for expression outlier analysis using RNA-seq Outlier Detection in Cancer (RODIC)[27], ESTIMATE [28] for immunological gene signatures/infiltrates and ssGSEA [29] for pathway analysis. Quality control metrics are captured in Shiny QC and reviewed. For calling variants, BAM files will be preprocessed similar to exome methods, except an additional trimming of soft-clipped reads is performed prior to indel re-alignment and base recalibration. HaplotypeCaller [8]will generate variant call files (VCF), prior to variant effect predictor analysis of annotated mutations [7]. Tophat Fusion Detection v2.0.10 [30], STAR-fusion v1.4.0 [31] and MAVIS [32] will be run for detection of fusion candidates. MiXCR v2.1.10 [33] will be applied for analysis of tumor T and B cell receptor repertoire. Clinical protocols that allow for data sharing as defined in patient consent forms, or as approved by ethical review boards may use a secured instance of cBioportal [20, 21] for visualization of genomic data cohorts.

Clinical reporting for CAP transcriptome assays are limited to fusion gene detection and percentile and outlier gene expression (see figure 6.4b). All CAP transcriptomes will be annotated on genome build hg38.p12.

PM-OICR TGL

Page 19 of 28 Version 1.9

Figure 6.4a: Tumor transcriptome pipeline.

Figure 6.4b: CAP Tumor transcriptome pipeline.

6.5 Methylation Sequencing Informatics

PM-OICR TGL

Page 20 of 28 Version 1.9

Sequence reads from Illumina TruSeq Methylation EPIC exome libraries will be aligned using Bismark software v0.15.0 [34]. Methyl-converted BAM files will be processed using Bismark Methyl Extractor generating bed graph files of % CpG methylation for each site. Library quality metrics including mapping efficiency, % reads on target, and distribution of % CpG methylation across samples will be calculated. Clinical protocols that allow for data sharing as defined in patient consent forms, or as approved by ethical review boards may use a secured instance of cBioportal [20, 21] for visualization of genomic data cohorts.

Figure 6.5a: Tumor methylation pipeline.

6.6 cfMeDIP Informatics

For pre-analytic MiSeq QC analysis and library validation, unique molecular indexes (UMIs) are trimmed using trimmomatic v0.33[35] prior to FastQC analysis [5]. Trimmed reads are aligned against hg19 human reference using BowTie v 2.1.0 to generate raw sequence alignments in BAM format [6]. CollectMultipleMetrics, CollectGcBiasMetrics, and MarkDuplicates are run via Picard tools suite v2.4.1[36].

PM-OICR TGL

Page 21 of 28 Version 1.9

Deep sequenced (NextSeq550 or NOVA) cfMeDIP libraries are analyzed as above and QC metrics captured. Untrimmed reads containing UMIs are processed with ConsensusCruncher [37] to suppress sequencer errors; duplicate reads are amalgamated into single-strand consensus sequences and combined into duplex consensus sequences. Singletons (reads lacking duplicate sequences) are corrected and combined with single-strand consensus sequences and collapsed to unique molecules. De-duplicated reads are aligned using BwaMem v 0.7.15 [6] and analyzed with the MEDIPS R package (v. 1.22.0) [38]. Counts bed files are generated for windows of 100bp, 200bp and 300bp.

cfMeDIP IP libraries are assessed for CpG relH and CpG GoGe enrichment, AT dropout, and methylated Arabdopsis enrichment relative to control/pre-capture library. Density clustering by ctDNA methylation status using a t-Distributed Stochastic Neighbor Embedding (t-SNE) method may be applied. Tumor content/fraction and large scale copy number alterations are predicted using ichorCNA [24] from control/pre-capture library as in figure 6.3a.

(a) Figure 6.6a: cfMeDIPs workflow.

6.7 ctDNA Informatics

Sequence reads are analyzed with FastQC [5] and aligned against hg19 human reference using BwaMem v 0.7.12 to generate raw sequence alignments in BAM format [6]. Using ConsensusCruncher [37], unique molecular indexes (UMIs) from the sequencing library are utilized to suppress sequencer errors; duplicate reads are amalgamated into single-strand consensus sequences and combined into duplex consensus sequences. Singletons (reads lacking duplicate sequences) are corrected and combined with single-strand consensus sequences and collapsed to unique molecules. Mutations are called with MuTect1 v1.1.7 [9] and Strelka v1.0.13 [10]. Pre-capture libraries may be used to assess tumor content/fraction and large scale copy number alterations using ichorCNA [24] as in figure 6.3a.

6.8 Infinium Methylation EPIC Array Informatics

IDAT files generated from the Illumina iScan array scanner will be preprocessed using Bioconductor package minfi 1.2 [39]. Data will be normalized with ssNoob. A correction for tissue type (FFPE/frozen) may be performed by using the removeBatchEffect function of the

PM-OICR TGL

Page 22 of 28 Version 1.9

Limma package version 3.30.9 [40]. M-values will be calculated based on log 2 ratios of the intensities of methylated versus unmethylated probes prior to annotation. Signature discovery may use a combination of Spectral Clustering (SNF), Multidimensional Scaling Plot (MDS), Hierarchical clustering, consensus clustering, machine learning or other methods to achieve a stable clustering of a signature probe set. 7. Contracts: Material and Data Transfer Agreements, Statement of Work Forms

7.1 Collaborations between UHN & PM-OICR TGL, Statement of Work Forms

PM-OICR TGL is a joint collaborative initiative partially funded by OICR and UHN (Princess Margaret Cancer Foundation). The master collaboration agreement MCA 2016-0694 defines intellectual property ownership for all projects initiated between UHN and PM-OICR TGL; a copy is available via the UHN Technology Development and Commercialization (UHN TDC) Office. This agreement accelerates and streamlines inter-institute workflows facilitating rapid project initiation and information exchange.

All UHN investigators with projects originating within UHN and collaborating with PM-OICR TGL must complete a statement of work form (SOW) available on our website. A completed SOW is a requirement of the UHN-OICR master agreement MCA2016-0694. Investigators or designates, or PM-OICR TGL’s project manager may draft a statement of work (SOW). The SOW outlines the project, budget, materials, data, and IP ownership or IP study clauses/conditions. The initiating investigator must approve the SOW, which is then authorized by the Deputy Director of OICR, Dr. Christine Williams. The PM-OICR TGL project manager registers the completed SOW with UHN TDC and UHN TDC assigns a UHN contract ID. All parties receive a copy of the executed SOW. The project may launch after UHN registration, pending CAPCR approval of the study protocol. See section 8.1 for further details on ethics compliance. Ethics and SOW documentation will be registered with OICR’s Research Compliance Officer.

7.2 Collaborations between PM-OICR TGL and an external institute

Projects that initiate external to UHN that do not involve UHN scientists (with the exception of Dr. Trevor Pugh, PM-OICR TGL Director, UHN employee) require an external institute or OICR material transfer agreement. UHN is a party to the agreement, and may request changes to the agreement. The project manager of PM-OICR TGL will file an administrative ethics review with the University of Toronto ethics board for single study site protocols from Toronto Academic Health Sciences Network (TAHSN) affiliated ethics boards. Multi-site clinical protocols, or non-TAHSN approved protocols require a delegated ethics review by the University of Toronto ethics review board. See section 8.2 for further details on ethics compliance. Ethics and material transfer agreements will be registered with OICR’s Research Compliance Officer. Material transfer agreements will also be registered with UHN TDC.

PM-OICR TGL

Page 23 of 28 Version 1.9

8. Ethics

The Tri Council Policy Statement: Ethical Conduct of Research Involving Humans (TCPS2) [41] permits the appointment of multiple ethics boards for protocols conducted at PM-OICR TGL, including UHN’s Coordinated Approval Process for Clinical Research (CAPCR) ethics board or OICR’s board of record, the University of Toronto My Research Human Protocols (MHRP) ethics board. TCPS 2 specifies that “an official agreement clarify the ultimate responsibility of the institution for the ethical acceptability of research undertaken within its jurisdiction or under its auspices” must exist (application of article 6.1). OICR and UHN have established this agreement.

8.1 UHN CAPCR Approved Protocols

UHN collaborators are required to complete a statement of work form (available on our website) which is registered at UHN’s Technology Development and Commercialization (TDC) office (refer to subsection 7.1). All UHN initiated projects must have CAPCR approval prior to sample receipt at PM-OICR TGL, and are within CAPCR board jurisdiction; they do not require ethics approval by the University of Toronto Research Ethics Board. PM-OICR TGL requires your UHN CAPCR approval letter, study protocol, consent templates and project statement of work, and continuing approval letters. Please forward all amended documents and amendment approvals to the PM-OICR TGL project manager. These documents will be registered with OICR’s Research Compliance Officer.

8.2 External Ethics Board Approved Protocols (Non-CAPCR)

Protocols with ethics approval(s) from external boards require review by the University of Toronto Research Ethics Board (MHRP). The PM-OICR TGL project manager will file an administrative ethics review with the University of Toronto ethics board for single study site protocols from Toronto Academic Health Sciences Network (TAHSN) affiliated ethics boards. Multi-site clinical protocols, or non-TAHSN ethics board approved protocols require a delegated ethics review by the University of Toronto ethics review board. PM-OICR TGL requires your institutional review board’s (iRB) approval letter, study protocol, consent templates and continuing approval letters. Forward all protocol amendment documentation to the PM-OICR TGL project manager for amendment with the University of Toronto Research Ethics Board. University of Toronto Research Ethics Board approval is contingent on an executed research agreement. Contracts and study documentation will be registered with OICR’s Research and

PM-OICR TGL

Page 24 of 28 Version 1.9

Compliance Officer. UHN will receive a copy of the executed research agreement which will be registered with UHN TDC.

9. Publication Policy

9.1 Authorship Policy

PM-OICR TGL, UHN Tumor Immunotherapy Program, and the UHN Cancer Genomics Program follow authorship guidelines from the International Committee of Medical Journal Editors (ICJME, http://www.icmje.org/icmje-recommendations.pdf). Authorship credit should be based on 1) substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; 2) drafting the article or revising it critically for important intellectual content; and 3) final approval of the version to be published. Authors should meet conditions 1, 2, and 3.

When a large, multicenter group has conducted the work, the group should identify the individuals who accept direct responsibility for the manuscript. These individuals should fully meet the criteria for authorship/contributorship defined above, and editors will ask these individuals to complete journal-specific author and conflict-of-interest disclosure forms. When submitting a manuscript authored by a group, the corresponding author should clearly indicate the preferred citation and identify all individual authors as well as the group name. Journals generally list other members of the group in the Acknowledgments. The NLM indexes the group name and the names of individuals the group has identified as being directly responsible for the manuscript; it also lists the names of collaborators if they are listed in Acknowledgments.

All persons designated as authors should qualify for authorship, and all those who qualify should be listed. Each author should have participated sufficiently in the work to take public responsibility for appropriate portions of the content. Acquisition of funding, collection of data, or general supervision of the research group alone does not constitute authorship. In addition to the above, the following general principles apply: leadership role, concept origin, group principles to foster collaborations and promotion of new/young investigators.

(a) Suggested Authorship Listing for Investigator Initiated Trial (Assuming 20 authors):

• 1st author (up to 3 co-first authors): Trainee (e.g. clinical or pathology fellow, graduate student, or post doctoral fellow) or lab associate actively involved in project (when appropriate)

• 2nd author: Top accruing study investigator or fellow (cannot be trial PI or coPI) • 3rd author: Top accruing study pathologist (if applicable) based on number of cases

reviewed for the study

PM-OICR TGL

Page 25 of 28 Version 1.9

• 4th – 9th authors (6 spots): Remaining trial investigators or fellows based on accrual • 10th – 15th authors (6 spots): Remaining trial immunophenotyping, genomics, or other lab

representatives based on internal agreement • 16th author: trial study coordinator or correlative studies coordinator (alternate in

different papers) • 17th author: trial clinical or scientific PI (may serve as co-corresponding author) • Last author: trial clinical or scientific PI (corresponding author)

9.2 Publication Acknowledgements

All individuals who played a contributing role to the trial, including accrual, sample collection and analysis, will be included in an Acknowledgements section (unless already listed as authors).

For investigator initiated studies including genomic data, the following text should be included (or modified if individuals included have been credited with authorships):

“We thank the staff of the Translational Genomics Laboratory (https://labs.oicr.on.ca/translational-genomics-laboratory) for their expertise in generating and analyzing the sequencing data used in this study. The Translational Genomics Laboratory is a joint initiative between the Princess Margaret Cancer Centre and the Ontario Institute for Cancer Research that is enabled through funding provided by the Government of Ontario, Ministry of Research, Innovation and Science and the Princess Margaret Cancer Foundation.”

10. Revision History

Version Number

Date (yyyy-mm-dd) History of change

1.0 2017-08-18 Template document created (DT) 1.1 2017-08-30 Added buffy coat/plasma isolation, ctDNA

isolation, low pass whole genome library prep 1.2 2017-09-08 Updated Cibersort/Immunomap information 1.3 2018-02-16 Added protocols for cfMeDIP, ctDNA. Added

Illumina NovaSeq platform to protocols. (Dax Torti)

1.4 2018-02-27 Updated exome informatics with tumor only methods including CNV kit, and global mutational signatures method with deconstructSigs. Removed certain tools from

PM-OICR TGL

Page 26 of 28 Version 1.9

RNA Informatics. Updated references. (Jon Torchia/Dax Torti)

1.5 2018-03-20 Included shipping address and procedures, added low pass whole genome informatics process. (Dax Torti)

1.6 2018-07-04 Added recommended core clinical data elements and added guidance on de-identification of patient direct identifiers, updated exome and transcriptome informatic pipeline documentation. (Dax Torti)

1.7 2018-08-15 Updated contract and ethics processes. (Dax Torti)

1.8 2018-08-21 Added fresh frozen tissue embedding, added guidance on collection of blood, ctDNA, cfMeDIP informatics. Added publication and acknowledgement policies.

1.9 2019-06-17 Added updated WGS,shallow WGS guidance, references for Consensus Cruncher, guidance on EDTA vs Streck, and ctDNA input reference documentation. Updated figures to higher resolution.

2019-11-20 Added guidance on ctDNA LOD and panel validation. Updated image storage details to integrate Azure storage.

11. References:

1. Conley, R.B., et al., Core Clinical Data Elements for Cancer Genomic Repositories: A Multi-stakeholder Consensus. Cell, 2017. 171(5): p. 982-986.

2. Risberg, B., et al., Effects of Collection and Processing Procedures on Plasma Circulating Cell-Free DNA from Cancer Patients. J Mol Diagn, 2018. 20(6): p. 883-892.

3. Kis, O., et al., Circulating tumour DNA sequence analysis as an alternative to multiple myeloma bone marrow aspirates. Nat Commun, 2017. 8: p. 15086.

4. Wang, L., S. Wang, and W. Li, RSeQC: quality control of RNA-seq experiments. Bioinformatics, 2012. 28(16): p. 2184-5.

5. Andrews, S., FastQC. 2018. 6. Li, H. and R. Durbin, Fast and accurate short read alignment with Burrows-Wheeler transform.

Bioinformatics, 2009. 25(14): p. 1754-60. 7. McKenna, A., et al., The Genome Analysis Toolkit: a MapReduce framework for analyzing next-

generation DNA sequencing data. Genome Res, 2010. 20(9): p. 1297-303. 8. Poplin, R., et al., Scaling accurate genetic variant discovery to tens of thousands of samples.

bioRxiv, 2017. 9. Cibulskis, K., et al., Sensitive detection of somatic point mutations in impure and heterogeneous

cancer samples. Nat Biotechnol, 2013. 31(3): p. 213-9.

PM-OICR TGL

Page 27 of 28 Version 1.9

10. Saunders, C.T., et al., Strelka: accurate somatic small-variant calling from sequenced tumor-normal sample pairs. Bioinformatics, 2012. 28(14): p. 1811-7.

11. Talevich, E., et al., CNVkit: Genome-Wide Copy Number Detection and Visualization from Targeted DNA Sequencing. PLoS Comput Biol, 2016. 12(4): p. e1004873.

12. McLaren, W., et al., The Ensembl Variant Effect Predictor. Genome Biol, 2016. 17(1): p. 122. 13. Lek, M., et al., Analysis of protein-coding genetic variation in 60,706 humans. Nature, 2016.

536(7616): p. 285-91. 14. Chang, M.T., et al., Identifying recurrent mutations in cancer reveals widespread lineage diversity

and mutational specificity. Nat Biotechnol, 2016. 34(2): p. 155-63. 15. Memorial Sloan Kettering Cancer Centre, Q.D. OncoKB. 2016; Available from: Oncokb.org. 16. Landrum, M.J., et al., ClinVar: public archive of interpretations of clinically relevant variants.

Nucleic Acids Res, 2016. 44(D1): p. D862-8. 17. Alexandrov, L.B., et al., Signatures of mutational processes in human cancer. Nature, 2013.

500(7463): p. 415-21. 18. Rosenthal, R., et al., DeconstructSigs: delineating mutational processes in single tumors

distinguishes DNA repair deficiencies and patterns of carcinoma evolution. Genome Biol, 2016. 17: p. 31.

19. Favero, F., et al., Sequenza: allele-specific copy number and mutation profiles from tumor sequencing data. Ann Oncol, 2015. 26(1): p. 64-70.

20. Cerami, E., et al., The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov, 2012. 2(5): p. 401-4.

21. Gao, J., et al., Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal, 2013. 6(269): p. pl1.

22. Quinlan, A.R. and I.M. Hall, BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics, 2010. 26(6): p. 841-2.

23. Scheinin, I., et al., DNA copy number analysis of fresh and formalin-fixed specimens by shallow whole-genome sequencing with identification and exclusion of problematic regions in the genome assembly. Genome Res, 2014. 24(12): p. 2022-32.

24. Adalsteinsson, V.A., et al., Scalable whole-exome sequencing of cell-free DNA reveals high concordance with metastatic tumors. Nat Commun, 2017. 8(1): p. 1324.

25. Dobin, A., et al., STAR: ultrafast universal RNA-seq aligner. Bioinformatics, 2013. 29(1): p. 15-21. 26. Li, B. and C.N. Dewey, RSEM: accurate transcript quantification from RNA-Seq data with or

without a reference genome. BMC Bioinformatics, 2011. 12: p. 323. 27. Torchia, J., RNA Seq Outlier Detection (RODIC). 2018. 28. Yoshihara, K., et al., Inferring tumour purity and stromal and immune cell admixture from

expression data. Nat Commun, 2013. 4: p. 2612. 29. Subramanian, A., et al., Gene set enrichment analysis: a knowledge-based approach for

interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A, 2005. 102(43): p. 15545-50.

30. Kim, D. and S.L. Salzberg, TopHat-Fusion: an algorithm for discovery of novel fusion transcripts. Genome Biol, 2011. 12(8): p. R72.

31. Haas, B., et al., STAR-Fusion: Fast and Accurate Fusion Transcript Detection from RNA-Seq. bioRxiv, 2017.

32. Reisle, C., et al., MAVIS: merging, annotation, validation, and illustration of structural variants. Bioinformatics, 2019. 35(3): p. 515-517.

PM-OICR TGL

Page 28 of 28 Version 1.9

33. Bolotin, D.A., et al., MiXCR: software for comprehensive adaptive immunity profiling. Nature Methods, 2015. 12: p. 380.

34. Krueger, F. and S.R. Andrews, Bismark: a flexible aligner and methylation caller for Bisulfite-Seq applications. Bioinformatics, 2011. 27(11): p. 1571-2.

35. Bolger, A.M., M. Lohse, and B. Usadel, Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics, 2014. 30(15): p. 2114-20.

36. Picard Tools. http://broadinstitute.github.io/picard. 37. Wang, T.T., et al., High efficiency error suppression for accurate detection of low-frequency

variants. Nucleic Acids Res, 2019. 38. Lienhard, M., et al., MEDIPS: genome-wide differential coverage analysis of sequencing data

derived from DNA enrichment experiments. Bioinformatics, 2014. 30(2): p. 284-6. 39. Fortin, J.P., T.J. Triche, Jr., and K.D. Hansen, Preprocessing, normalization and integration of the

Illumina HumanMethylationEPIC array with minfi. Bioinformatics, 2017. 33(4): p. 558-560. 40. Ritchie, M.E., et al., limma powers differential expression analyses for RNA-sequencing and

microarray studies. Nucleic Acids Res, 2015. 43(7): p. e47. 41. Canadian Institutes of Health Research, N.S.a.E.R.C.o.C. and a.S.S.a.H.R.C.o. Canada., Tri-Council

Policy Statement: Ethical Conduct for Research Involving Humans, December 2014. 2014.