Tweet
Hello,
I am absolutely new to the area and am doing my self-education based on web resources. I try to establish a pipeline for bacterial RNA-seq and sRNA discovery based on Illumina NGS data. I primarily use R as my working language; this area seems to be at an early stage and Bioconductor packages are scattered. I've already written my own script to count genome coverage at a single-nucleotide level (leveraging functionality of ShortRead and Biostrings packages). I've also tried a conventional route from FASTQ to BAM files using Bowtie + samtools. As a result, here I am: start from FASTQ => Bowtie => SAMtools => BAM files; I also have my single-nucleotide coverage matrices which seem to match the information in the SAM/bam files.
Here is a point where I suddenly stuck:
Conversion of a "raw" genome coverage to coverage based on genomic features. I tried to read docs to GenomicRanges, IRanges etc packages but cannot see a clear path through them.
What is your favorite way from BAM (using genome annotation .gff files) to a matrix of a feature-centric coverage ready to be used for differential expression tests? And again, the genome has nothing to do with Homo sapiens.
Thanks a lot!
I am absolutely new to the area and am doing my self-education based on web resources. I try to establish a pipeline for bacterial RNA-seq and sRNA discovery based on Illumina NGS data. I primarily use R as my working language; this area seems to be at an early stage and Bioconductor packages are scattered. I've already written my own script to count genome coverage at a single-nucleotide level (leveraging functionality of ShortRead and Biostrings packages). I've also tried a conventional route from FASTQ to BAM files using Bowtie + samtools. As a result, here I am: start from FASTQ => Bowtie => SAMtools => BAM files; I also have my single-nucleotide coverage matrices which seem to match the information in the SAM/bam files.
Here is a point where I suddenly stuck:
Conversion of a "raw" genome coverage to coverage based on genomic features. I tried to read docs to GenomicRanges, IRanges etc packages but cannot see a clear path through them.
What is your favorite way from BAM (using genome annotation .gff files) to a matrix of a feature-centric coverage ready to be used for differential expression tests? And again, the genome has nothing to do with Homo sapiens.
Thanks a lot!
Comment