Skip to main content

Common Bioinformatics Data Types and Formats

Understanding the various file formats used in bioinformatics is essential for effective data management, analysis, and reproducibility. This guide covers the standard formats for sequences, alignments, variants, and genomic features.

🧬 Sequence Formats​

FASTA and FASTQ files are the foundational formats for next-generation sequencing (NGS). They represent the starting point for most bioinformatics pipelines, carrying the raw or processed code of life.

FASTA​

A FASTA file can contain multiple nucleotide or amino acid sequences. Each entry starts with a header line beginning with the > symbol, followed by a unique identifier and metadata (e.g., gene name, organism). The subsequent lines contain the actual sequence data.

  • Note: FASTA files typically do not contain information about the quality of the sequence.

FASTQ​

A FASTQ file is an extension of the FASTA format that also stores corresponding quality scores. Each read entry is recorded in four lines:

  1. Sequence identifier
  2. Nucleotide sequence
  3. A + delimiter
  4. Quality scores (encoded as ASCII characters) indicating the confidence of each nucleotide call.

📍 Alignment Formats​

After sequencing, a major task is "mapping" or aligning these reads to a reference genome to determine their origin.

SAM (Sequence Alignment Map)​

The Sequence Alignment Map (SAM) format is generic text format for storing the alignments of sequencing reads against reference nucleotide sequences and is typically the output of read alignment algorithms, such as Bowtie2. Each file contains one header section and alignment section with the read sequence and quality scores such as the reference chromosome, genomic position, mapping quality, and a CIGAR string describing matches/insertions/deletions. The SAM format supports short and long reads (up to 128 Mbp) produced by different sequencing platforms.

BAM (Binary Alignment Map)​

The Binary Alignment Map (BAM) format is a compressed, binary format of SAM and thus take up less storage space than SAM files (usually 30-50% smaller). Some special tools are needed in order to make sense of BAM, such as Samtools, Picard, and Integrative Genomics Viewer (IGV).

The detailed format specification a the SAM/BAM formats can be found in the official web site: https://samtools.github.io/hts-specs/.

STO (Stockholm)​

The STO (Stockholm) format is a text file format for storing multiple sequence alignments (MSAs) and additional alignment features and metadata. They typically have .sto or .stk file extensions. The Stockholm format is used by the Pfam and Rfam databases and tools like HMMER. A STO filE consists of a header line with a format and version identifier; mark-up lines starting with "#=GF","#=GC","#=GS" or "#=GR"; alignment lines with the sequence name and aligned sequence; a "//" line indicating the end of the alignment. Alignments are shown with inserts as lower case characters, matches as upper case characters, and gaps as ' . ' or ' - '.

A more detailed description of Stockholm files can be found here: https://rfam.xfam.org/help#tabview=tab13.

🔍 Variants​

VCF (Variant Call Format)​

The VCF (Variant Call Format) is a tab-delimited text file format originally developed under the 1000 Genomes Project and is the standard output of variant callers. It is used to store information sequence variations in the genome, such as Single Nucleotide Polymorphisms (SNPs) and insertions/deletions (indels). It contains meta-information lines (prefixed with ##), a header line (prefixed with #), and data lines each containing 8 fields for each variant in the genome, typically followed by additional columns showing format and sample information.

BCF (Binary Call Format)​

The Binary Call Format (BCF) is the binary version of the VCF format, allowing for faster processing, reading, and storage for large projects. Tools like BCFtools can work with both VCF and BCF files and are often used to convert between them.

More detailed information on the VCF and BCF files can be found here: https://samtools.github.io/hts-specs/VCFv4.2.pdf.

📊 Counts and Expression Matrices​

For transcriptomics (RNA-seq), data is often represented as an expression matrix—a table where rows represent genes and columns represent samples or individual cells.

  • Text Files (TSV/CSV): Simple tabular formats used for smaller datasets.
  • The Hierarchal Data File (HDF5) is a file format designed to store large amounts of omics data and metadata. HDF5 uses a hierarchical directory structure that act as containers to store two types of objects: 1) groups, containing zero or more groups; and 2) datasets, multi-dimensional arrays of read counts together with supporting metadata. This format allows for storing both the count matrices and all metadata in a single file rather than having separate features, barcodes and matrix files. This makes it suitable for single-cell sequencing data.

🗺️ Gene Feature Formats​

These formats define the "map" of the genome, indicating where genes, exons, and introns are located.

  • The General Feature Format (GFF) format is a standard, general-purpose format for storing genomic annotations. A GFF file generally contains the name of the sequence, the type of feature described (i.e., a whole gene, an exon or an intron), the coordinates of the feature within the overall sequence and other additional attributes. It contains genes and other features of DNA, RNA and protein sequences, each line representing a single feature and containing nine fields: seqid, source, type, start, end, score, strand, phase, and attributes. The most recent version of GFF is GFF3. The official documentation for the GFF format can be found here: https://github.com/The-Sequence-Ontology/Specifications/blob/master/gff3.md

  • GTF (Gene Transfer Format): A variation of GFF specifically focused on gene structures.

  • The Browser Extensible Data (BED) is a tab-separated file used to store basic genomic features. BED files are typically generated by feature detection algorithms or through manual curation of alignments. Each entry consists of one line per feature, with the first three required columns specifying the genomic region interval (contig/chromsome name, begin, and end position); the remaining nine columns contain additional track information.

    The full BED specification can be found here: http://genome.ucsc.edu/FAQ/FAQformat.html#format1.