Accel-Amplicon™ 56G Oncology Panel v2

Optimal coverage uniformity, sensitivity, and specificity are achieved with qPCR-verified DNA input amounts in the 10-25 ng range. Between 25-100 ng, coverage uniformity may be mildly reduced while sensitivity and specificity are preserved. Using less than 10 ng may reduce specificity of the assay and affect variant calling for low frequency alleles due to low copy number.

Accel-Amplicon panels perform best within the 10-25 ng input range, so it is important to have an accurate understanding of the input material. Quantification by a fluorometric assay such as Qubit is satisfactory for high quality DNA from whole blood, fresh frozen, or cultured cells. However, this method may significantly overestimate the amplifiable content of a more fragmented sample, such as FFPE or cfDNA, because it reflects double-stranded content regardless of molecule size. A qPCR-based assay can be designed such that the primers will only produce amplicons from fragments larger than a given size, which is a better indicator for how well they will amplify in typically compromised samples such as FFPE and cfDNA. The table below illustrates the variability that can be expected across 10 different FFPE samples when quantified with three common methods, including the absorbance-based NanoDrop, the fluorescence-based Qubit, and a qPCR assay. Please note that the Input DNA Quantification Assay provides full details for the published ALU115-qPCR assay and there are also commercial options for qPCR input quantification available.

The level of multiplexing is directly correlated to the size of the specific Accel-Amplicon Panel and the flow cell chemistry utilized for sequencing of the libraries. The table below provides an example of the number of libraries from the Accel-Amplicon Comprehensive TP53 and 56G Oncology Panels which can be multiplexed on a single MiSeq^® flow cell. Accel-Amplicon panels currently include up to 96 indexing combinations and are consistently increasing.

*Please inquire by email to Tech Support if you require more than 96 unique indices.

Lower than expected yields most likely relate to the quality and quantity of the sample, as well as how it was quantified. If possible, for severely compromised samples including FFPE, use 25 ng of qPCR-quantified input and extend the incubation time for the Indexing Step from 20 minutes to 60 minutes to improve yields. It is very important to quantify cfDNA or FFPE with qPCR as opposed to Qubit or NanoDrop to ensure there is a minimum of 10 ng of amplifiable content in the sample.

It is not recommended to use a Bioanalyzer or Qubit for quantifying libraries because there is no PCR enrichment of the library following the Indexing Step, so the Bioanalyzer or Qubit will not accurately quantify fully adapted library vs. other DNA. In addition, the Accel-Amplicon library adapters have secondary structure which exhibit migration artifacts on the Bioanalyzer.

Low cluster density is typically related to an error in library quantification. If the final library is quantified by methods other than qPCR, this will lead to determining an inaccurate value for library concentration. It is not recommended to use a Bioanalyzer or Qubit for quantifying libraries because there is no PCR enrichment of the library following the Indexing Step, so the Bioanalyzer or Qubit will not accurately quantify fully adapted library vs. other DNA.

In addition, the Accel-Amplicon library adapters have secondary structure which exhibit migration artifacts on the Bioanalyzer. When diluting the library for loading the flow cell and sequencing, if the dilution is based off an erroneous quantification the cluster density will not be optimal.

Swift is pleased to accept custom Accel-Amplicon Panel requests. We will use a custom design pipeline to generate the primer pools and provide a functionally-tested pilot kit, with pricing decided on a case-by-case basis dependent on reaction volumes and assay complexity. Please contact us by email at custom@swiftbiosci.com and specify:

• Number of genes to cover, with gene symbols
• Hotspot (i.e., SNP) or whole-gene coverage requirements
• Sample type(s) to be used with the custom panel such as FFPE, cfDNA, or high quality gDNA
• Expected number of reactions required for the custom panel

There are multiple factors which determine the LOD, the most important being the number of copies of the variant-containing DNA actually present in the sample and the depth of sequencing performed.

There is also some variation in the reasonable LOD depending on the variant of interest. In general, the Accel-Amplicon technology is capable of a LOD down to 1% for most base substitution variants when working with 10-25 ng input DNA as quantified by a qPCR input assay.

It is very important to quantify cfDNA or FFPE with qPCR as opposed to Qubit or Nanodrop to ensure there is a minimum of 10 ng of amplifiable content in the sample for this LOD to be achieved, since for 10 ng this will be represented by ~30 mutant copies detected in ~3000 total copies.

It is also critical to achieve adequate sequencing depth to obtain sufficient mutant copy number detection. The table below illustrates observable allele frequencies for Accel-Amplicon libraries at 10 ng and 1 ng input amounts.

Please note that Swift Biosciences currently supports inputs down to 10 ng.

For maximal (lowest) LOD, increasing the input amount to 25 ng, if possible, will provide more confidence due to increased copy number ( > 6000 total copies and therefore > 60 copies of a 1% variant).

It is also important to note that the LOD inherent in Illumina technology can be discussed in terms of bases rated at Q30. Most bases are read with a rate of 1 error in 1000, or 0.1%. Some bases have a lower quality score and therefore the background false positive noise level is between 0.1-0.6%.

Using a base quality filter during the variant calling steps is one way to address this issue.

Swift does not currently offer a proprietary data analysis software package. Please view our technical note “Accel-Amplicon Panels: Bioinformatics Guidelines“, which includes a tutorial for using open-source Linux-based tools to perform the required primer trimming step, as well as general recommendations for the alignment and variant calling analysis.

While this can be caused by multiple factors, one of the most common explanations is that the Sample Sheet was not set to automatically trim the Illumina adapter sequences when generating the FASTQs prior to the primer trimming requirement of Accel-Amplicon Panels.

Make sure that the “Use adapter trimming” and “Use adapter trimming Read 2” are selected during the sample sheet setup. It is possible to re-run this analysis even after the sequencing run has been performed if these selections were not made during the initial sample sheet setup.