Mapping: Answers

Mapping using HISAT2

Alignment summary

768 reads; of these: 768 (100.00%) were paired; of these: 94 (12.24%) aligned concordantly 0 times 674 (87.76%) aligned concordantly exactly 1 time 0 (0.00%) aligned concordantly >1 times ---- 94 pairs aligned concordantly 0 times; of these: 12 (12.77%) aligned discordantly 1 time ---- 82 pairs aligned 0 times concordantly or discordantly; of these: 164 mates make up the pairs; of these: 104 (63.41%) aligned 0 times 59 (35.98%) aligned exactly 1 time 1 (0.61%) aligned >1 times 93.23% overall alignment rate

How many RNA-seq read pairs were provided as input to HISAT2?

768 pairs (768*2 = 1536 reads)

From HISAT2 summary output file, or input FASTQ file.

How many of those read pairs were mapped by HISAT2?

1432 = ((674+12)*2) + (59+1)

How many reads were uniquely mapped, i.e. mapped to one genomic location?

1431

This can compute this by running:

grep -cw 'NH:i:1' hisat2.sam

Or if you have made a BAM file:

samtools view hisat2.bam | grep -cw 'NH:i:1'

Or by looking at the summary output file, summing up:

2 * 674 concordantly and uniquely mapped pairs 2 * 12 discordantly and uniquely mapped pairs 59 uniquely mapped reads lacking a mapping for the paired read

In general, do the alignments seem to be good? I.e. do they cover the entire reads and contain few mismatches?

Yes.

Look at mapping quality, CIGAR string and NM tag (edit distance) in the SAM file:

cut -f5-6,12- hisat2.sam | more

Or for the BAM file:

samtools view hisat2.bam | cut -f5-6,12- | more

Mapping using STAR

Alignment summary

                                Started job on |       Oct 22 22:45:41
                            Started mapping on |       Oct 22 22:47:03
                                   `Finished on |       Oct 22 22:47:06
      Mapping speed, Million of reads per hour |       0.92

                         Number of input reads |       768
                     Average input read length |       202
                                   UNIQUE READS:
                  Uniquely mapped reads number |       733
                       Uniquely mapped reads % |       95.44%
                         Average mapped length |       199.98
                      Number of splices: Total |       81
           Number of splices: Annotated (sjdb) |       68
                      Number of splices: GT/AG |       81
                      Number of splices: GC/AG |       0
                      Number of splices: AT/AC |       0
              Number of splices: Non-canonical |       0
                     Mismatch rate per base, % |       0.26%
                        Deletion rate per base |       0.00%
                       Deletion average length |       0.00
                       Insertion rate per base |       0.00%
                      Insertion average length |       1.33
                            MULTI-MAPPING READS:
       Number of reads mapped to multiple loci |       5
            % of reads mapped to multiple loci |       0.65%
       Number of reads mapped to too many loci |       0
            % of reads mapped to too many loci |       0.00%
                                 UNMAPPED READS:
      % of reads unmapped: too many mismatches |       0.00%
                % of reads unmapped: too short |       3.91%
                    % of reads unmapped: other |       0.00%
                                 CHIMERIC READS:
                      Number of chimeric reads |       0
                           % of chimeric reads |       0.00%

How many RNA-seq read pairs were provided as input to STAR?

768 pairs (768*2 = 1536 reads)

from Log.final.out, or input FASTQ file

How many of those read pairs were mapped by STAR?

738, sum of 733 uniquely mapped 5 multimapped

from Log.final.out

How many reads were uniquely mapped, i.e. mapped to one genomic location?

1466

This can be obtained from Log.final.out (2 * 733 uniqely mapped reads).

Or by running:

grep -cw 'NH:i:1' Aligned.out.sam

Or on a BAM file:

samtools view star.bam | grep -cw 'NH:i:1'

In general, do the alignments seem to be good? I.e. do they cover the entire reads and contain few mismatches?

Yes.

See Log.final.out, in particular Average mapped length and Mismatch rate per base. Can also use samtools (see the answer to the corresponding question for HISAT above).

Converting SAM to BAM

Does the output from samtools flagstat confirm any of your answers to the questions in the HISAT2 and STAR sections above?

Kind of. Note that several of the numbers refer to the number of mappings (not reads) and there can be several mappings per read.

Load the the BAM files with HISAT2 and STAR results into IGV. Go to the RAB11FIP5 locus. Have HISAT2 and STAR mapped the reads in a similar way?

Yes.

Detailed examination of the read alignments in IGV should indicate if the RNA-seq data is strand-specific. Is it?

No.

Right-click on the track name, choose Color alignments by > First of pair strand. If the data is strand-specific, one color should dominate for every gene (depending on the strand of the gene).