ExpansionHunter Denovo (EHdn) can be used to analyze a collection of BAM/CRAM files containing alignments of short (100-200bp) reads. For best results, the samples should be sequenced on the same instrument to similar coverage of at least 30x. All data should be aligned with the same short-read aligner ideally without any post-processing steps such as indel realignment or recalibration.
Your dataset should contain one or more samples that are suspected to harbor a repeat expansion and a set of controls. If the controls are not available, consider using Illumina Polaris dataset.
EHdn supports two secondary analysis types: case/control analysis and outlier analysis. Case/control analysis is appropriate for detecting expansions present in a significant proportion of cases and largely absent from controls. Outlier analysis, on the other hand, should be used when an expansion might be present only in one or few case samples.
Example 1 If the set of cases consists of many related individuals exhibiting the same phenotype then a case/control analysis may be appropriate.
Example 2 If the set of cases consists of ALS patients, the case/control analysis is expected to flag the GGCCCC repeat in C9orf72 gene as highly significant.
Example 3 If the set of cases is an undiagnosed rare disease cohort exhibiting a mix of phenotypes, it is unlikely that there is an enrichment for any specific expansion and hence the outlier analysis is appropriate.