to
Close
close

Member Login


User ID  
Password

Wet lab

1. Introduction

Next generation sequencing (NGS) has been adopted in all areas of molecular diagnosis. Laboratories must become familiar with the critical differences between NGS and traditional Sanger sequencing. The wet laboratory process is one such area of critical difference. Robust quality assurance and quality control procedures are essential to ensuring the reliability of NGS testing results.

This chapter will focus on the “wet” laboratory issues including laboratory environment, sample/library preparation, template generation, sequencing and quality assurance in genomic diagnostic application.

1.1 Wet lab processes

Many of the guidelines in this document are common to all forms of nucleic acid testing. These guidelines should be read in conjunction with ISO 15189 and all relevant NPAAC documents, but particularly Requirements for Medical Testing of Human Nucleic Acids, and Requirements for the Development and Use of In-House In Vitro diagnostic Devices.

We propose that these principles and guidance could form a foundation for future specifications of performance and formal regulations of genomic testing. It is not our intention to generate a user guide and provide all the solutions. Instead, we try to include some relevant resources for your reference. For example, some relevant “wet” laboratory issues can be found from the website of the Division of Laboratory Programs, Standards, and Services (DLPSS) of the American Centers for Disease Control and Prevention (CDC).

1.1.1 Resources

2. Measures to Control Contamination

2.1 The laboratory should be designed to minimise the contamination of samples at different stages of the workflow with other specimens or amplified products.

Laboratories should ensure the physical design can accommodate separate areas for patient derived samples and amplified material.

Possible cross contamination between these areas including by movement of equipment, staff, or aerosols should be assessed and managed.

Measures should be available to both detect cross contamination between clinical samples, and to eliminate it. Detection may include the use of processing blanks or environmental monitoring. Elimination may include the use of hypochlorite or other decontamination measures.

For further information refer to refer to NPAAC Requirements for the Medical Testing of Human Nucleic Acids.

2.2 Cross contamination between samples due to carryover from equipment:

Laboratories should ensure recommended and appropriate maintenance and cleaning processes are performed to eliminate carryover contamination.

Laboratories should include a monitoring process for carryover contamination as part of regular internal quality control.

Sample indexes (barcodes) used to identify unique reads in pooled libraries can be used to detect carryover contamination. These should be re-used on the longest cycle possible. Consecutive runs of the same sequencing instrument using the same barcode indexes should be avoided. Frequent re-use of the same set of barcode indexes will compromise the laboratory's ability to detect cross-contamination at any stage of the sequencing procedure.

2.3 Sample Indexing should be performed at the earliest possible stage of library preparation to allow subsequent detection of cross-contamination.

The laboratory should avoid workflows that offer the potential for undetectable sample cross-contamination. Workflows that call for multiple manipulations, additions, and incubations of samples prior to index ligation or amplification increase the risk of undetectable sample to sample cross-contamination whereas workflows which add unique indexes to each sample early in the library preparation process provide a means to make cross-contamination detectable.

2.4 Laboratories should consider including identity SNPs within the assay to confirm patient identity.

Identity SNPs’ could be included within each assay and interrogated with a second method to confirm patient identity, if no unique variants are identified within the genes analysed. These SNPs can also be used to monitor and detect any carryover contamination within the data. Where members of the same pedigree have been analysed, bioinformatics analyses to confirm family relatedness may also prove useful to highlight errors in specimen identification, processing or contamination.

3. Wet Workflow Validation

3.1 The genomic platform used must meet the specifications required for the diagnostic purpose and be operated in accordance with best practice as determined by the manufacturer.

Consideration should be given to biases inherent in the platform of choice. Particular attention should be given to ensuring that any systematic weaknesses or errors of the sequencing system do not limit the diagnostic specificity of the assay, or that if such flaws exist, that orthogonal testing is employed to detect variants in regions of bias. Examples include regions of high GC content or repetitive regions.

3.2 Diagnostic laboratories should validate the operational performance of the wet laboratory workflow used in molecular diagnosis.

Expansion of genomic methods for diagnostic applications makes it increasingly important to demonstrate data quality, reliability and reproducibility. Diagnostic laboratories should empirically determine their minimum requirements for data quality.

Analytic sensitivity and specificity are important performance characteristics for genomic diagnostic applications. Diagnostic laboratories should document these aspects of the laboratory workflow by comparison of test results obtained under conditions defined above, to those obtained from a gold standard method (usually Sanger sequencing).

3.2.1 Resources

3.3 Diagnostic laboratories should regularly monitor the performance of the wet laboratory workflow used in molecular diagnosis.

Inclusion of known DNA control/standard samples at <10% of the pooled libraries at regular intervals would allow ongoing monitoring of assay performance and data analysis processes.

3.4 The use of outsourced platforms and services for diagnostic services should meet all of the standards outlined in this document

If part of the genomic testing process is to be outsourced, NATA accredited providers or providers showing full compliance with NPAAC standards must be used. It remains the responsibility of the clinical laboratory to review, retain and furnish for audit all documentation related to clinical testing.

4. Sample Preparation

4.1 The laboratory should assess the quantity and quality of DNA samples before proceeding with diagnostic application.

Failure to exclude samples of poor quality or insufficient quantity of amplifiable DNA can significantly affect the sensitivity and specificity of genomic diagnosis and lead to the possibility of false negative results. This is of particular significance where the sample type may be associated with limiting amounts of DNA, for example FFPE tissue or cell-free circulating DNA. Failure of sample exclusion can also affect turnaround time, due to the long cycle of the genomic testing process.

In the case of measuring cell-free circulating DNA for the purposes of non-invasive prenatal screening or testing, the laboratory should have a process to ensure that adequate amounts of foetal DNA (i.e. in accordance with the sensitivity limit determined for the assay) are present in the sample prior to data analysis and interpretation of results.

4.2 Diagnostic laboratories should determine an appropriate range of DNA sample concentration and types to be included for an efficient test using genomic methods.

Where appropriate, consideration should be given to including related affected and unaffected samples in the analysis. For example sequencing trios (proband and both parents) to confirm a de novo change, or tumour and normal samples to exclude a cancer variant as germline.

4.3 When only small amounts of tissue are available for somatic testing, the laboratory should determine the minimum specimen size and tumour proportion needed for successful analysis.

Assessment of tissue volume and cellularity is usually estimated by microscopic examination by a competent person. Sufficient purity or proportion of targeted cells can then be achieved through macro-dissection.

5. Library Preparation

5.1 The laboratory should have an effective system to track the samples during the multiple-step process of library preparation.

For laboratories handling in excess of 1000 samples per year, a Laboratory Information Management System capable of tracking a multistep workflow, with multiple samples, and QC steps should be considered.

5.2 The laboratory should have a quality control procedure to assess the adequacy of DNA fragmentation procedures.

For those laboratories that use protocols making use of DNA fragmentation, quality assessment of DNA fragmentation procedure is essential to ensure the right size distribution and accurate amount of fragmented DNA samples. The latter is critical for equal molar representation if multiple barcoded samples are to be subsequently pooled for library preparation.

5.3 The laboratory should undertake quality assurance measures during the validation phase to demonstrate that no significant allele bias or allele dropout occurs during target enrichment processes.

The laboratory should determine the optimal conditions for library preparation. Documented metrics of performance of library preparation should be generated and used to QC library preparation steps on all clinical samples. For example, effect of input mass of DNA, fragmentation conditions, PCR cycles, etc. should be assessed. QC metrics in the form of Bioanalyser traces, spectrophotometric readings, or real-time PCR results should be produced and routinely collected and compared to those of an optimal validated run.

6. Template Generation

6.1 The laboratory should have a quality assessment procedure to assess the quality and quantity of a prepared DNA library used for template generation.

An accurate estimation of DNA library quantity is essential for optimal clonal amplification.

Quantification should be based on amplifiable templates (i.e. DNA fragments with proper ligated adaptors). For example, quantitative PCR (qPCR) has high levels of sensitivity and specificity and can accurately measure quantities of DNA.

6.2 The laboratory should have a quality assessment procedure to assess the adequacy of clonal amplification used for template generation.

Quality assessment of the clonal amplification procedure is essential to ensure an adequate representation of DNA samples in the template. This is critical for equal representation if multiple barcoded samples have been pooled during library preparation.

7. Data Generation

7.1 The laboratory should establish empirically the coverage necessary for accurate detection of sequence variants and copy number changes, and provide the best estimation of false positive and negative rates.

The laboratory should employ quality control measures that specify the quantity and quality of DNA sequence data to accurately differentiate all targeted sequence variants. This is especially critical when a multiplexed target enrichment procedure has been used to generate libraries

The laboratory should ensure that there is sufficient coverage for the detection of aneuploidy e.g. in non-invasive prenatal trisomy 21 testing.

7.2 If multiple samples are to be sequenced simultaneously, the laboratory should have quality assurance measures to demonstrate that DNA sequence data generated cannot be attributed to the wrong sample.

Consideration should be given to the use of barcoded DNA samples and the possibility of sequence data being misdirected to the wrong specimen.

7.2.1 Resources

7.3 Data should be stored as required for diagnostic DNA studies.

Consideration should be given what would be the suitable data format to keep (see further discussion in Bioinformatics Section). The raw reads and quality scores should be kept as a minimal requirement.

Data storage should also comply with overarching regulatory and legislative requirements (see section in Ethical & Legal Issues.)

7.4 Any exception should be recorded for patient samples where steps used in the analytical process deviate from laboratory standard operating procedures.

This exception log should be kept with the reason(s) for deviation and should retain links to the patient sample.

7.4.1 Resources

8. Quality Control and Quality Assurance

Consideration should be given to cross platform confirmation. Sanger sequencing should be considered to reduce false positive and/or negative rates, particularly in small indel variants.

The limitation of genomic testing should be presented in the final report (See the details in the Reporting section).

QC of sequencing data may include:

  • Base call quality scores
  • Read depth
  • Uniformity of read coverage
  • Read enrichment (for capture-based methods)
  • Percentage PCR duplicates (for capture-based methods)
  • Allelic Read Percentage
  • GC bias
  • Decline in signal intensity along a read

8.2 The laboratory should implement quality assurance measures that evaluate the entire process.

Well-characterised DNA samples should be used as internal quality control samples. Cell lines are renewable, but may have some balanced or unbalanced chromosomal rearrangements. Blood samples from young subjects (<55 years) are typically free from such rearrangements, but have limited supply. Rearrangements that are identified may reflect the age of the donor or be a consequence of the culture process. Consideration should also be given to obtaining reference materials from overseas. For example, the Food and Drug Administration of the United States of American has recently completed the Sequencing Quality Control (SEQC) project, as a part of Phase III of the MicroArray Quality Control (MAQC-III) project. Its aims were to assess the technical performance of genomic platforms by generating benchmark datasets with reference samples, and to evaluate the advantages and limitations of various bioinformatics strategies in RNA and DNA analyses.

Acceptable intra-and inter-run variability should be established during validation and monitored in diagnostic laboratories. It is important to determine assay precision, i.e., the degree to which repeated measurements give the same result – both repeatability (within-run precision) and reproducibility (between-run precision).

Genomic technologies are rapidly evolving. Consideration should be given whether positive findings in genomic analysis should be confirmed by a different chemistry or a second method, particularly at the initial validation stage and for results that affect clinical decision-making.

The laboratory should monitor, implement and validate upgrades to instruments, sequencing chemistries and reagents or kit used to generate genomic data.

8.2.1 Resources

8.3 Laboratories performing diagnostic genomic testing should participate in suitable genomic proficiency testing or inter-laboratory sample exchange programs to meet the requirements for external quality assessment measures.

Laboratories should establish a reportable range for each assay, such as multiple genes, exome and large genomic regions.

8.3.1 Resources

Contact form for feedback and comments:

 

*Required

*Required

*Required

 Security code