cufflinks problem: one gene have multiple rpkm values

[calliopsis]

Hi,

I got several rpkm values for one gene (gene.expr file, not tranascript) when I use cufflinks. I trimmed my reads because of bad quality score. I run Tophat for different lengthes separately. Then I combined all the different length of result (accepted.sam file) and sorted them according to the start site for each chromosome, after that I run cufflinks, I found I got several rpkm values for one gene. Does anyone can explain this and how to solve it?

[PFS]

I had similar problem. Apparently there is a bug in cufflinks that will be fixed soon.