How to evaluate reads quality of a sample for a bacterial genome

[nivea]

Dear all,

After mapping the reads to a reference genome, is there a good standard to evaluate the mapped reads for a sample?
For example, if the whole transcriptome is 4,000 kbp, what's the minimum number of uniquely mapping reads to cover this transcriptome?
If there're so many gaps (regions without reads mapped) within some genes, can we say the reads are good enough?
For a single gene, can we be sure to say it is expressed if its average base coverage is 0.5?

Thanks!