RNA-seq reads counting matrix

[doublealice]

I have a list of reference genome loci coordinates. The short RNA reads were mapped on these loci by bowtie and samtools.

Instead of simply get the counting number of reads, I hope to generate a matrix like this:

read-id   Loci1     Loci2     Loci3 .....
  1          2            4         0
  2          0            3         4
  3          4            2         5
...

This is just an example showing that read1 aligned two different places in Loci1, 4 different places in loci2, none in loci3.

I had made loci table in a gff format file and loaded it into R. All reads and alignment information is in bam file.

I tried two ways to load bam file into R. One is:

reads<-readBamGappedAlignments("myalign_reads.sorted.bam")

I found the output ignore my reads ID. Only chromosome and positions and other information are available. I need reads ID to count how many different places a read hits within one loci. This seems to be a limitation of the package.

I also tried another method:

indexBam("myalign_reads.sorted.bam"0
param<-ScanBamParam()
reads<-scanBam("myalign_reads.sorted.bam", index="myalign_reads.sorted.bam", param=param)

This loaded everything and read id was stored in qname. However, I don't know how to perform the further counting to generate the matrix.

Can you give me any suggestion or idea on how to work on it? I tried my best but due to limit knowledge on R and bioconductor, I have suffered from it for several days.

I appreciate if anyone can give me some hints or help me work it out. Thanks!

Alice

[areyes]

Sorry, I did not watch your matrix correctly!

[doublealice]

I got some progress on this according to the manual from http://manuals.bioinformatics.ucr.edu/home/ht-seq
on section "Computing Absolute and Relative Overlaps Among Ranges".

Now, I have a list like this:

readid     loci-id
1           105
1            108
3            114
3            114
2            120
4            123
3            123
5            124
5            125
5            125
6            128
7            128
6            128
6            128
8            128
...

The next work is transform this table into my expected matrix, like:

readid   105 108  114  120  123  124  125  128
1          1    1
2                          1
3                    2             1
4                                       1
5                                               1      2
6                                                        3
7                                                    1
8                                                   1

I don't know how to adjust the table to present it very well here. Anyway, it just shows how many different places a read mapped in a loci.

To get this matrix also not easy for me. If you know any R function can work on it, please hint. Thank you very much!

Alice

=========
I got it. Using table() can get that matrix.