Variability Correction Informatics Baseline (VCIB) baseline workflow presets

In Ion AmpliSeq™, Ion AmpliSeq™ HD, and Oncomine™ assays, somatic copy number estimates are made when Ion Reporter™ Software counts reads for each amplicon, and makes adjustments to account for specific types of variability. The software then compares the read counts to expected counts for the amplicons and then makes further adjustments in a set of normal samples. That is, for plasma or cell-line samples that do not have known CNVs, or in baseline samples with no known copy number variants.

Known sources of variability include pool imbalance (when the assay has more than one pool of amplicons), total number of reads and, for each amplicon, attributes of GC proportion, and length of the amplicon insert. In practice, we observe other variability that does not associate with known attributes yet is systematic. The method that is used by the software to identify copy number variations is based on many diverse samples, captures systematic effects, and encodes the samples into a file that contains the baseline information.

You can create a Variability Correction Informatics Baseline (VCIB) to detect copy number variation in samples. You can either create a baseline and use that baseline as is, or you can augment an existing baseline with samples. The VCIB baseline must include a minimum of 48 samples, and at least 6 of the samples must be designated as normal. All samples that are used to create the baseline must have been analyzed in Ion Reporter™ Software with the same custom panel.

Runs included in the baseline should cover the technical variability that is present in the research laboratory, such as the number of PCR cycles, multiple Ion Chef™ Instruments if present, laboratory technicians, variations of protocols, and so on. Typically a VCIB baseline is created for one chip type, for example, the Ion 550™ Chip.

For capturing sample to sample variability, it is recommended to select a cohort of diverse sample types. Including multiple samples of every sample type to which the CNV baseline are applied. This always includes clinical research samples such as FFPE. It can also include control samples, cell lines, and so on.

For FFPE samples, it is recommended to select samples with no known CNV aberrations in the region covered by the panel, if possible. If this information is not available, it is recommended that more than 48 distinct libraries be used with no more than 2 library replicates.

When a baseline is augmented with samples, new samples are run, the size of each systematic effect that is encoded in the baseline is estimated, and a correction is applied to remove the effect. The added samples must be diverse to capture likely systematic variation.

Note: If you use a single sample analysis workflow, you need to use a copy number baseline to detect CNVs. You need to create a custom baseline if a baseline is not shipped with a custom panel.

To use a VCIB baseline in Ion Reporter™ Software, you can select the baseline preset when you create an analysis workflow. For more information, see Apply a copy number baseline workflow preset to an analysis workflow.