Good bowtie2 alignment rate and low mapping on transrate ...

Question

[doinelpierrot]

Hello everybody !

I've run my first assembly recently and I'm trying to figure out its quality. I have honest n50 and a not so bad BUSCO of 80% complete busco. However, I am receiving mixed signals from Transrate and Bowtie2.

For both of them I have sent as input my trimmed fastq files and my final Trinity fasta. However, I got a 98 % overall alignment rate with bowtie2 but a p fragment mapped of 0.22 on Transrate that comes with a ridiculously low assembly score of 0.02...

I really want to say it is contradictory and forget the transrate thing, but it really makes me doubt .

Here is mybowtie2 output :

215513959 reads; of these:
  215513959 (100.00%) were paired; of these:
    33155696 (15.38%) aligned concordantly 0 times
    47425197 (22.01%) aligned concordantly exactly 1 time
    134933066 (62.61%) aligned concordantly >1 times
    ----
    33155696 pairs aligned concordantly 0 times; of these:
      2974711 (8.97%) aligned discordantly 1 time*
 30180985 pairs aligned 0 times concordantly or discordantly; of these:
      60361970 mates make up the pairs; of these:
        7085190 (11.74%) aligned 0 times
        6679869 (11.07%) aligned exactly 1 time
        46596911 (77.20%) aligned >1 times
98.36% overall alignment rate

Thank for your help

[JC]

134933066 (62.61%) aligned concordantly >1 times 46596911 (77.20%) aligned >1 times

Most of your alignments fall into multiple locations, so, your reads are too short or you have a lot of repetitive sequences

[doinelpierrot]

Ok so I have many reads that map different places of my transcriptom. I know I had a lot of duplicates in my reads. Could it be an explanation ?