Hello, 

 I used a Tophat 2 analysis and got 99.2% discordant in the summary, while in fact I checked        the distance between the reads  and got 170 with std of 35 (this is the data i filled in the              Tophat2).

  

Why did the Tophat2 test found 99.2% discordant when it should have been much lower?

The starting reads was 16665209, in a Bowtie2 test i got a logical  outcome of 16567482, while on the other hand i got from Tophat2 35066991 and 34816158 - not logical. 

I tried to do the same test number of times with another data  and got resemblance in the results. 

In all of the tests in Tophat2 i got discordant 99% and higher. 

What am i doing wrong? 
What is the correct way to feed data into the test? 

Thank you

Si.