HGTStart v5 takes a FASTA file containing exactly one protein sequence (plus its genome FASTA, annotation file and NCBI taxonomy id) to predict HGTs.

Notified by email

Input required
One TARGET protein per job — put it as the FIRST record. You may include its neighboring proteins from the same genome as additional records for flanking-gene context. Content after the first space in ID will be ignored.
Example: XP_015873120.1 (histone-like HGT gene of Ziziphus jujuba) + chromosome proteins
Provide the annotation of your genome. GFF3 / GTF / GFF are accepted; please make sure protein IDs in the annotation match the FASTA header.
Example: chromosome NC_029680.1 annotation (14 MB)
Example: chromosome NC_029680.1 sequence (28 MB)
Example: 326968 (Ziziphus jujuba)

Optional parameters

Search & trees (steps 1-2)
DIAMOND search engine (fixed)

Homolog search & tree building
Homolog sampling and search parameters are managed automatically by the pipeline (DIAMOND search, IQ-TREE / FastTree tree building).
HGT inference (step 3)
Donor(s); separate multiple donors with comma
node support Cutoff, 0-100 (70 recommended; 0 = skip)
Optional taxa allowed in monophyletic ingroup
taxa to be Ignored while screening trees
minimal Strongly Supported Nodes uniting query and donors(default=1)
minimal number of All Supporting Nodes uniting query and donors(default=2)
minimal number of Receptors in a nested position (default =2)
minimal number of Donors in a nested position (default =2)
minimal OutGroup Size for a tree to be considered valid(default=5)

HGT inference filters
minimal value of Alien Index to filter HGT events (small value = keep more candidates)
minimal value of Branch Length Index to filter HGT events (small value = keep more candidates)
minimal value of BLAST Score Index to filter HGT events (small value = keep more candidates)

Taxon resolution & genomic context
coding loci per side of an HGT block (default=1)