Many diseases esp. neurological diseases are caused by tandem repeat expansion, for example trinucleotide/hexanucleotide repeat, though pathogenic mechanism is still not fully clear. And I think it's extremely challenging to look for such short repeat expansion from sequencing data; basically such expansion can be regarded as novel insertion. For large novel insertion, based on read alignment, we may find the breakpoint but may never figure out the exact inserted sequences (because it's too large); and another approach is to assemble sequences to find out large novel insertions.
Above is just my own thinking, what programs/algorithms would you recommend for identification of repeat expansion from sequencing data?