Downloads

Aligned FASTA for the full set of 496 named orthogroups, or just the 224 core proteins shown on the site (those with more than 10 sequences). AA is the amino acid alignment; CDS is the corresponding nucleotide codon alignment (same columns, back-translated from the AA alignment), for codon-based analyses such as dN/dS and codon-aware phylogenetics.

A CheckM-compatible database for assessing the completeness and contamination of plastid metagenome-assembled genomes. It bundles the profile HMMs, the lineage-specific marker-gene sets and the curation tree needed to place a bin in the taxonomy. It is built to be used with ChloroScan, a workflow dedicated to recovering plastids from metagenomes. Scripts to compute MAG completeness and contamination from the database manually are in the code repository.

Profile hidden Markov models (HMMs) for the 496 named orthogroups, in HMMER3 format, for homology searches. The core orthogroups (those with more than 10 sequences) additionally carry calibrated noise and trusted score cutoffs for reliable detection.

Raw Newick tree with 865 leaves: 857 algal taxa, one land-plant reference (Saxifraga limprichtii), and 7 cyanobacterial outgroups. Built from 51 conserved marker genes.

Orthogroups whose members are mainly hypothetical open reading frames (ORFs), with non-canonical names. AA is the aligned protein FASTA; CDS is the corresponding nucleotide codon alignment.

HHsuite results for the probability of 989 Pfam domains being present in 470 proteins, tab-delimited.

Predicted structures for the 224 core proteins in mmCIF format, modelled with AlphaFold 3 from the orthogroup alignments. Subject to the AlphaFold 3 terms of use (see the note below).