Dingxia, It looks like TSS fits your need. Best regards, Julie On 8/29/12 1:04 PM, "Dingxia Feng" <fengdxia@iastate.edu> wrote: Hi, Julie, I'm going to use the worm database ws220 (I use this code to load ws220: wormbase=useMart("WS220")) to find the nearest genes and the distance to the translation start site of the nearest genes. The thing is c.elegans's transcripts will be trans-spliced, so it's hard to define the start of the transcription, so people usually use the start of the gene coding region. I check the help for the feature type you listed, they are "TSS","miRNA", "Exon", "5utr", "3utr", "ExonPlusUtr", "transcript" I'm not quite sure about your definition for "Exon" and "transcript". And I don't know if I use ws220 instead of ensemble, whether the definition for these feature types are the same. Would you please give some suggestion for my case to use ws220 to locate the peaks with the nearest start of the gene coding region? Thanks. Best. [[alternative HTML version deleted]]