Bacterial RNA seq analysis

[ateeqrr]

Dear all
I am analyzing bacterial transcriptome (illumina paired end reads) via online galaxy server.
I have uploaded fasta file of my reference genome to my account, and i am using it as reference for assembly.
my pipeline is Bowtie --> Cufflink --> cuffmerge --> cuffdiff
after cuffdiff, i am getting differentially expressed transcripts and differentially expressed genes. Number of lines in both files are same. does it mean number of genes are equal to number of transcripts.
some of the transcripts are of about 5Kb, i am sure it may have more than one gene. i would like to analyze my data based on genes rather than these big size locus.
does anyone knows, how i could achive it
Thanks in advance

[shriram]

You need to provide the gtf/gff file to tophat and cufflinks to get the known genes structure + novel genes.

[Jean]

This may not be relevant since the topic is old, but the cufflinks model is not useful for bacterial RNAseq analysis because of it's isoform-based assumptions. You would want to use something along the lines of edgeR or DEseq.