- Proper error catching for shape errors, issues with FASTA files, etc
- Test fully if the poly-A works. Extend to proteins perhaps?
- Refinement program
- Do not run segmentation on masked out areas
- Switch poly-A nucleotide to poly-U (or any pyrimidine)
- Cache sequence KV