Hi mswierniak,
The best and simplest metrics you can use (until someone proves otherwise) are number of reads of support, and number of subjects/independent experiments where you observe the same circle. When people say "unique reads" I believe they are referring to reads that do not have 100% identical sequence, because these are more likely to be amplification products. (Edit: In practice, these are usually called by looking for reads aligning to the same location in the genome.)
Your last question on expression is quite difficult, in my opinion. This recent paper may be of interest to you. http://genomebiology.com/2014/15/7/409
The best and simplest metrics you can use (until someone proves otherwise) are number of reads of support, and number of subjects/independent experiments where you observe the same circle. When people say "unique reads" I believe they are referring to reads that do not have 100% identical sequence, because these are more likely to be amplification products. (Edit: In practice, these are usually called by looking for reads aligning to the same location in the genome.)
Your last question on expression is quite difficult, in my opinion. This recent paper may be of interest to you. http://genomebiology.com/2014/15/7/409
Comment