Normalization giving the wrong species

[duncster94]

Hi, thanks for providing this tool!

I ran BERN2 on an example abstract from a paper on yeast RNA Pol III function (PMID: 10473618). The NER is generally very accurate however the normalization step maps each mention to a human gene, rather than a yeast gene.

I have a curated list of yeast gene function papers, so I know the majority of gene mentions in any given paper will be yeast related. Is it possible to specify a dictionary consisting of only yeast genes to avoid incorrect mapping to homologous genes in different organisms? Failing this, is it possible to only run the NER step using BERN2 and take care of the normalization myself?

Thanks!