genomeblocks.genes

Table of contents

1. Transcript(Locus)
2. Gene(Locus)
3. Genes(dict)
   1. Factories
   2. Properties
   3. Methods

---

Transcript(Locus)

@dataclass
class Transcript(Locus):
    transcript_id: str = ""
    exons: Loci
    cds:   Loci
    utr:   Loci

Methods: add_exon(e), add_cds(c), add_utr(u).

---

Gene(Locus)

@dataclass
class Gene(Locus):
    gene_id: str = ""
    gene_name: str = ""
    gene_type: str | None = ""
    transcripts: dict[str, Transcript]
    # computed in __post_init__:
    tss: Locus   # TSS as a 1-bp Locus (strand-aware)

Method: add_transript(t_id, t).

---

Genes(dict)

Subclass of dict[str, Gene].

Factories

Genes.make(filename, gene_name_key="gene_name",
                     gene_type_key="gene_type",
                     chr_map=None,
                     promoter_r=1000) -> Genes
# GTF/GFF parser. Handles gene/transcript/exon/CDS/five_prime_UTR/three_prime_UTR/UTR.

Genes.make_ucsc(filename, chr_map=None,
                          promoter_r=1000,
                          keep_alt_contigs=False) -> Genes
# UCSC RefSeq table parser (column order: bin, name, chrom, strand, txStart, ...).

Properties

Name	Description
filename	Source file path.
_promoter_r	Promoter half-window (bp).
annot	Lazy {'body','prom','exon','utr5','utr3'} dict of merged/sorted Loci.

Methods

Genes.get_tss(gene_type=None) -> dict[str, Locus]
# {gene_name → TSS Locus}; optional gene_type filter.

Genes.annotations(loci) -> pandas.DataFrame
# Columns: uid, annotation ∈ {Promoter-TSS, 5UTR, 3UTR, Exonic, Intronic, Intergenic}

Genes.nearest_genes(loci) -> pandas.DataFrame
# pyranges.nearest against slopped-TSS Loci; Name_b = gene name, Distance in bp.

Genes.table() -> str
# Summary of counts.