TODO_SGSeq_notes.md

• SGSeq: https://bioconductor.org/packages/3.16/bioc/vignettes/SGSeq/inst/doc/SGSeq.html

• plotCoverage() shows coverage by exon, and arcs per splice

• plotSpliceGraph() plots schematic splice graph, flattened per gene.

• TxFeatures class: importTranscripts() from GFF/GTF file, or GRangesList

  • code converts TxDb using convertToTxFeatures()

  • this function converts exons to GRanges

  • it creates junction regions between exons, also GRanges

  • exons have one or more txName, (CharacterList), and one or more geneName (CharacterList, maybe multiple are allowed?)

  • exons have a type:

    • J (splice junction)
    • I (internal exon)
    • F (first/5'-terminal exon)
    • L (last/5'-terminal exon)
    • U (unspliced transcript).

• SGFeatures class:

  • TxFeatures can be converted with convertToSGFeatures()

  • model of disjoint exons spliced into transcripts

  • exons have a type:

    • J (splice junction) - this junction could be used in splicejam
    • E (disjoint exon bin) - this exon could be used in splicejam
    • D (splice donor site) - this point feature is not used in splicejam
    • A (splice acceptor site) - this point feature is not used in splicejam

  • featureID unique identifier

  • txName CharacterList of transcripts; geneName CharacterList of genes

• Input typically by BAM file, STAR or GSNAP that custom tag XS for spliced reads

  • si (sample information) is a data.frame with these columns:

    • sample_name: character unique sample label
    • file_bam: character BAM alignment file
    • paired_end: logical indicating whether reads were paired
    • read_length: integer length of each read
    • frag_length: integer fragment length
    • lib_size: integer total aligned fragments

  • getBamInfo() can be used to create this data.frame:

    • path <- system.file("extdata", package = "SGSeq")
    • si$file_bam <- file.path(path, "bams", si$file_bam)