- Yield jobs to start downloading from the beginning
- Saving already downloaded files: Create a downloaded files register to avoid downloading multiple time the same file.
- Extend NCBI downloads to other data type than genomes
- Extend ENA downloads to other data types than raw reads
- NCBI: Use the output of datasets on wrong accession number to remove the bad accessions only (See how it is recursively done in SRA)
- Change the delay functions to hold an amount of query per second, not a fix delay
- Add url support for s3
- Add url support for gs
- Allow download path hierarchy