DHFR Protein Sequence Alignment Generated by HHblits

HHBlits DHFR alignment with gaps present is a collection of dihydrofolate reductase (DHFR) protein sequences. The dataset, created by Francois D. Rouleau, contains sequences with less than 70% homology to PjDHFR to avoid skewing GEMME analysis with highly similar sequences. It was last updated on 2026-05-27 and is available in FASTA format.

Use Cases

• Training protein language models based on curated DHFR sequences.
• Benchmarking multiple sequence alignment algorithms using the provided FASTA alignment.
• Studying protein evolution and homology within the DHFR family based on the described sequence diversity.

Strengths

• Dataset size is 2.0 MB, indicating a manageable download and processing footprint.
• Sequences are curated to have less than 70% homology to a reference (PjDHFR), which likely reduces redundancy.
• Data is provided in the standard FASTA format for biological sequences.

Limitations

• Row count is unknown, which may limit suitability assessment for large-scale training.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare, author Francois D. Rouleau

Collection Method

Aligned using HHblits, a tool for sensitive protein sequence searching.

Freshness

Last updated 2026-05-27 17:44:15; freshness should be verified.