Tumor-normal pair research

The predefined analysis workflows for tumor-normal research generate analysis results that identify reads of both a tumor sample and reads of the related normal sample. These predefined analysis workflows are optimized to find somatic variants, which appear in the tumor sample and do not appear in the normal sample. These predefined analysis workflows also perform a statistical evaluation of the likelihood that the tumor allele is not present in the normal sample and calculates a P value that represents the statistical confidence of that call.

At each position within a variant in the tumor research sample, the evidence for that allele in the normal sample is examined as part of the analysis. If the tumor allele is detected in the reads of the normal research sample in levels that are higher than the error rate, it is not considered to be a tumor-specific (somatic) variant. Therefore the variant is rejected (not called).

Tumor variants in which the same position in the normal sample has no coverage, or has low coverage, are designated as nonconfident. Nonconfident variants are not assigned a p-value, and are flagged with NC-LC (nonconfident because of low-coverage) in the adjacent field in the output VCF file.

Some nonconfident variants receive the NC-LF (nonconfident because of low-frequency) flag instead of a p-value. This occurs with variants for which both of these conditions are true:

• The allele frequencies for the variant are <10% in both the tumor and normal sample.

• The variant has a nonzero allele frequency in the normal sample.

A Non-Confident variant call means that the variant might appear not only in the tumor sample, but also in the normal sample. This could indicate that either a germline variant or a systematic error is present in both samples.

In the VCF file, ./. means a no-call in the normal sample and 0/0 means a homozygous reference call.

The tumor-normal pair predefined analysis workflows are run on a pair of research samples from the same individual. Ideally both research samples are sequenced on the same chip.

Note: When you create a user-defined analysis workflow for an Ion AmpliSeq™ Exome tumor-normal pair analysis, it is recommended that you do one of the following to help ensure that the correct parameters that are applied:

1. Make a copy of the predefined analysis workflow or a user-defined analysis workflow for use with Ion AmpliSeq™ Exome tumor-normal pairs and edit any desired parameters. For more information, see Copy and edit an existing analysis workflow.

2. Use the predefined BED file to create a new user-defined analysis workflow. For more information, see Create a new analysis workflow.

It is recommended that you not import the Ion AmpliSeq™ Exome panel BED file through either the import function in Ion Reporter™ Software, or with a manual import.