Skip to content

FILE_FORMATS.md

Latest commit

 

History

History
102 lines (80 loc) · 4.36 KB

FILE_FORMATS.md

File metadata and controls

102 lines (80 loc) · 4.36 KB
Raw

VX Genome Viewer — File Formats

VX is read-mostly. It auto-detects input by extension and creates the appropriate track type. This document lists every supported format, the track type it maps to, and the output formats VX writes.

Input formats

Format Extensions Track type Notes
FASTA .fa, .fasta, .fna, .fa.gz Sequence Reference genome; the first FASTA you open creates a new group. Gzipped FASTA is supported transparently.
GTF .gtf Gene Ensembl / GENCODE format with gene_id, transcript_id, exon_number attributes. Full exon / CDS / UTR structure.
GFF / GFF3 .gff, .gff3 Gene Generic Feature Format v3. Same downstream rendering as GTF.
BAM .bam Alignment Requires a .bai index next to the BAM (auto-located). Indexed range queries; reads stream in by region.
SAM .sam Alignment Whole-file read. Use BAM for large files.
BED .bed Feature BED3, BED6, and BED12 (transcript) variants.
BigBed .bigbed, .bb Feature Self-indexed binary BED.
BedGraph .bedgraph, .bg Signal Continuous signal with explicit (chr, start, end, value) rows.
Wiggle .wig Signal fixedStep and variableStep both handled.
BigWig .bigwig, .bw Signal Self-indexed binary; the recommended format for large signal data.
VCF .vcf, .vcf.gz Variant Variant calls with quality, genotypes, and INFO annotations. Index .tbi is auto-located if present.
BCF .bcf Variant Binary VCF.
BEDPE .bedpe Interaction Paired genomic intervals for chromatin contacts.

Output formats

Format Extension Where it comes from
BED .bed Peak calling, interval operations, gap detection, generic interval analyses.
BedGraph .bedgraph Signal analyses (GC content, Z-score, signal difference, peak prominence).
TSV .tsv Quantification (RPKM, TPM, CPM, feature counts), variant tables, alignment statistics.
PNG .png Viewport snapshot.
JPEG .jpg, .jpeg Viewport snapshot.
TIFF .tiff Viewport snapshot.
SVG .svg Vector viewport export (one-shot, full-resolution).
MP4 .mp4 Video recording (H.264).
WebM .webm Video recording (VP9).
GIF .gif Video recording (animated).
FASTA .fa Sequence region export.
VXS .vxs Session file (JSON-based; round-trips all open files, viewport, track layout, bookmarks).

Chromosome naming

VX resolves the three common chromosome-naming conventions automatically:

Convention Example Typical sources
UCSC chr1, chrX, chrM GRCh38 FASTA from UCSC, many BAM headers
Ensembl 1, X, MT Ensembl GTF annotations
NCBI / RefSeq NC_000001.11 RefSeq assemblies

The internal ChromosomeAliasResolver lets you mix files using different conventions in the same group — when you open a BAM that calls it chr1 and a GTF that calls it 1, they line up correctly.

Indexing

VX is sensitive to indexes:

  • BAM requires .bai for the seekable random-access used by the async region fetcher. If you only have .bam, generate the index with samtools index your.bam.
  • VCF uses .tbi if present (much faster); otherwise it falls back to a full-file scan.
  • BigBed / BigWig are self-indexed — no sidecar needed.

Compression

  • .gz is supported for FASTA, GTF, GFF, BED, and VCF (where the format spec allows).
  • BAM/BCF/BigBed/BigWig are inherently compressed binary formats and are handled natively.

Format quirks worth knowing

  • GTF / GFF: VX expects transcript_id and gene_id attributes on exon / CDS / UTR records. Files missing transcript_id will still load but exons won't be grouped into transcripts.
  • BAM: subsampling kicks in when the visible window contains more than max_reads reads (default 500). The badge in the top-right of the alignment track tells you when this is active. Raise the cap via the track options popover.
  • VCF: per-variant INFO/FORMAT fields are not normalised; what's displayed in Active Mode is what the VCF says.
  • BigWig: zoom levels are precomputed in the file; VX picks the best zoom for the current viewport automatically.