prokka

Rapid prokaryotic genome annotation

---

File naming

The prokka format requires four files per MAG:

• <MAG-name>.fna
• <MAG-name>.ffn
• <MAG-name>.faa
• <MAG-name>.gff

info

FASTA files must have unambiguous file extensions that indicate what kind of sequence they hold. The accepted extensions are .fna, .ffn, .faa for contig, gene, and protein sequences, respectively.

File format

tip

For more information on the prokka output files, visit the prokka repository.

FASTA of contigs (.fna)

The file must follow the FASTA format standard, with an additional requirement for the heading: it must start with the contig name, then (optionally) a space followed by a description.

FASTA of genes (.ffn)

The file must follow the FASTA format standard, with an additional requirement for the heading: it must start with the gene name, then (optionally) a space followed by a description.

FASTA of proteins (.faa)

The file must follow the FASTA format standard, with an additional requirement for the heading: it must start with the protein name, then (optionally) a space followed by a description.

GFF (.gff or .gff3)

The file must follow the General Feature Format (GFF). It must have columns representing the following data, in that order and without a header:

Column name	Column obligatoriness	Data type	Data nullability
seqid	Mandatory	String	Not nullable
source	Mandatory	String	Nullable
type	Mandatory	String	Not nullable
start	Mandatory	Integer	Not nullable
end	Mandatory	Integer	Not nullable
score	Mandatory	Float	Nullable
strand	Mandatory	String	Nullable
phase	Mandatory	Integer	Nullable
attributes	Mandatory	String	Nullable

GFF entries gene fragments (e.g. CDS, exon, etc) must include either locus_tag or ID in their attributes column. This is what parsomics uses to link GFF entries to genes.

Mapping to database

FASTAFile

Original data	FASTAFile field
FASTA file path	path
FASTA file extension	sequence_type 1
FASTA file name	genome_key 2

FASTAEntry

Original data	FASTAEntry field
FASTA entry ID	sequence_name 3
FASTA entry Description	description
FASTA entry Sequence	sequence

GFFFile

Original data	GFFFile field
GFF file path	path
GFF file name	genome_key 4

GFFEntry

Original data	GFFEntry field
GFF entry seqid column	sequence_name
GFF entry source column	source_name
GFF entry type column	fragment_type
GFF entry start column	coord_start
GFF entry end column	coord_stop
GFF entry score column	score
GFF entry strand column	strand
GFF entry phase column	phase
GFF entry attributes column	attributes 5

Footnotes

1. .fna for SequenceType.CONTIG ("CONTIG"), .ffn for SequenceType.GENE ("GENE"), .faa for FragmentType.PROTEIN ("PROTEIN"). ↩

2. The MAG name in the FASTA file name is used to query the primary key of the corresponding genome in the database. ↩

3. The "ID" refers to the sequence name, not the primary key! To avoid confusion, primary keys in parsomics are named key, not id. ↩

4. The MAG name in the GFF file name is used to query the primary key of the corresponding genome in the database ↩

5. For easier access to data, this column is converted from a string to a JSONB property in the database. ↩