How to make a heatmap with the exons of a gene with bam file of RNA-seq in R-studio?

[César]

Hi, I´m new in this kind of analysis in RNA-seq but I´m trying to make a heatmap with the exons of a gene. My files are from a public database and first a convert the SRR file to bam file, After that I wanted to make my heatmap in R studio with that bam files but the R-studio console show me this messsage:
> # Crear el objeto DGEList
> dge <- DGEList(counts = matriz_conteo, genes = rownames(matriz_conteo))
Aviso:
In DGEList.default(counts = matriz_conteo, genes = rownames(matriz_conteo)) :
At least one library size is zero

After that, I had been trying to search what happen and apply the next codes to know the sizes of my files and if the position of the exons corresponds to the positions of the report exon positions in the reference genome that I used (GRCh38/hg38.p14). The size of my file is 0 and the position of my exons do not corresponds with the exon positions of the reference genome. I put the codes:

> dge$samples$lib.size
[1] 0 0 0 0 0 0 0 0


> exones_dmd
GRangesList object of length 1:
$ENSG00000198947
GRanges object with 275 ranges and 2 metadata columns:
seqnames ranges strand | exon_id exon_name
<Rle> <IRanges> <Rle> | <integer> <character>
[1] X 31097677-31098183 - | 1011488 ENSE00001871364
[2] X 31119219-31121930 - | 1011489 ENSE00001478643
[3] X 31119219-31121930 - | 1011490 ENSE00003916589
[4] X 31119221-31121930 - | 1011491 ENSE00001880031
[5] X 31119222-31121930 - | 1011492 ENSE00003901420
... ... ... ... . ... ...
[271] X 33211282-33211497 - | 1011761 ENSE00003920795
[272] X 33211282-33211519 - | 1011762 ENSE00003918256
[273] X 33211282-33211549 - | 1011763 ENSE00001899141
[274] X 33211282-33211554 - | 1011764 ENSE00001479071
[275] X 33339259-33339609 - | 1011765 ENSE00001478378
------

[fchatonnet]

Hi César,
it's not very clear how did you get from the bam files to the DGElist object, it seems that you used some kind of counts matrix, but how did you generate it?
Next, I'm not very familiar with using R to do this kind of analysis, but it would be faster to use bedtools, a command-line tool to manipulated bed (and bam) files) to achieve this. Using bedtools intersect -c with a bed file of your exons of interest as a first input and your bam file as a second input would directly count the reads which overlap the exons. Then you can you this count matrix in R to generate a heatmap.
Good luck!