RNA Seq Read Representation by Trinity Assembly

Assessing the Read Content of the Transcriptome Assembly

Assembled transcripts might not always fully represent properly paired-end reads, as some transcripts may be fragmented or short and only one fragment read of a pair may align. Simply aligning reads to your transcriptome assembly using bowtie or STAR will only capture the properly paired reads. To assess the read composition of our assembly, we want to capture and count all reads that map to our assembled transcripts, including the properly paired and those that are not.

In order to comprehensively capture read alignments, we run the process below. Bowtie2 is used to align the reads to the transcriptome and then we count the number of proper pairs and improper or orphan read alignments.

First, build a bowtie2 index for the transcriptome:

% bowtie2-build Trinity.fasta Trinity.fasta

Then perform the alignment to just capture the read alignment statistics.

Example for paired-end reads:

% bowtie2 -p 10 -q --no-unal -k 20 -x Trinity.fasta -1 reads_1.fq -2 reads_2.fq  \
     2>align_stats.txt| samtools view -@10 -Sb -o bowtie2.bam 

Example for single-end reads:

% bowtie2 -p 10 -q --no-unal -k 20 -x Trinity.fasta -U single.reads.fq \
     2>align_stats.txt| samtools view -@10 -Sb -o bowtie2.bam 

Visualize statistics:

% cat 2>&1 align_stats.txt

The output bellow:

76201190 reads; of these:
  76201190 (100.00%) were paired; of these:
    18166307 (23.84%) aligned concordantly 0 times
    17026716 (22.34%) aligned concordantly exactly 1 time
    41008167 (53.82%) aligned concordantly >1 times
    ----
    18166307 pairs aligned concordantly 0 times; of these:
      1769907 (9.74%) aligned discordantly 1 time
    ----
    16396400 pairs aligned 0 times concordantly or discordantly; of these:
      32792800 mates make up the pairs; of these:
        15287552 (46.62%) aligned 0 times
        3874965 (11.82%) aligned exactly 1 time
        13630283 (41.56%) aligned >1 times
89.97% overall alignment rate

A typical Trinity transcriptome assembly will have the vast majority of all reads mapping back to the assembly, and ~70-80% of the mapped fragments found mapped as proper pairs (yielding concordant alignments 1 or more times to the reconstructed transcriptome).

Note, if you see high rates of duplication, usually this is not a cause for concern. Highly expressed transcripts with alternatively spliced isoforms will account for multiple read mappings. Later, abundance estimation (transcript expression quantification) takes these multiple mappings into account.

Visualize read support using IGV

The Integrative Genomics Viewer is useful for visualizing read support across any of the Trinity assemblies. The bowtie2 alignments generated above, which are currently sorted by read name, can be re-sorted according to coordinate, indexed, and then viewed along with the Trinity assemblies using the IGV browser as follows.

Sort the alignments by coordinate

samtools sort bowtie2.bam -o bowtie2.coordSorted.bam

Index the coordinate-sorted bam file

samtools index bowtie2.coordSorted.bam

Index the Trinity.fasta file

samtools faidx Trinity.fasta

View the aligned reads along the Trinity assembly reference contigs.

note, you can do this by using the various graphical menu options in IGV (load genome 'Trinity.fasta', load file 'bowtie2.coordSorted.bam'), or you can use the command-line tool like so:

igv.sh -g Trinity.fasta  bowtie2.coordSorted.bam

Note, the above assumes that the Trinity.fasta and bowtie2.coordSorted.bam files are in your current working directory.

And you can then go to any Trinity assembly of interest and examine the read (and paired-end) support. An example region of a long Trinity transcript contig is shown below.