1) I've made tblast of a certain protein and have a list of alignments of my reference protein. 2) With accession numbers of these proteins i've download the coding sequences for them (through biopythons functions to get genbank format through accessions) and aligned them instead of proteins. 3) Knowing (through cruzdb) the annotation of these transcripts i can extract exons coordinates and find any region of interest inside transcript alignments, but i'm interested in finding the corresponding regions inside their genomes.

So my questions is actually twofold: first, is it possible to obtain genome regions through blast hits automatically? second one is about cruzdb functionality: I can download any genome region if a have a url:

twoBitToFa http://hgdownload.soe.ucsc.edu/goldenPath/mm10/bigZips/mm10.2bit -bed=input.bed stdout

But can i transform programmatically an accession number into a link to reference genome?

unfortunately, we are talking about fruit fly dm3 and also i need a command line tool - a web server would be hard to implement in a pipeline

It can, but do you know if it is possible to switch automatically to different organism/assemblies? It seems like the same problem here as with cruzdb - you can easily do it if you know the organism and assembly, but if you have only blast record - can you generate a url for reference?