In the past two days I’ve written a simple annotator program, that given an input list of RefSeq genes, automatically determines the relevant Entrez Gene IDs and annotates them using the flat files provided by the NCBI. A direct conversion was not possible due to limitations in Biopython’s parsers, but I managed to use the GenBank parser to identify and extract the references to the Gene IDs (and putting them in a list).

Once that had been done, I created a series of dictionaries when reading the annotation file, for data such as gene name, symbol, chromosome and cytoband. Using the list I already obtained, it was easy to create a new file with the required fields.

During this process I learnt somewhat more about how to play with iterators to skip headings and so on. The code is not yet sufficiently generic, but once I finish toying with it, I may publish it for “general” (assuming anyone would use it) consumption, under GPL v2.