This crate provides Rust types and parsers for the GFA (Graphical Fragment Assembly) format, version 1 and 2.
This library is compatible with the version 2 specification and version 1 specification of GFA.
This library it's a variation of the library developed by Christian Fischer link here.
This library performs 2 main operation on a file:
- Checking if a file is conform to a specified format and then create a GFA2 (or GFA) object.
Given a file for example: file.gfa2
H VN:Z:2.0
H ul:Z:https://github.com/pmelsted/GFA-spec/issues/7#issuecomment-219685552
S 11 5 ACCTT pg:J:{"Human":[{"target":"chr1","pos":1500,"strand":true}],"Mouse":[{"target":"chr2","pos":2000,"strand":false}],"ecoli":[{"target":"chr1","pos":2000,"strand":false},{"target":"chr1","pos":3000,"strand":true}]}
S 12 6 TCAAGG
S 13 7 CTTGATT
E * 11+ 12- 1 5$ 2 6$ 4M
E * 12- 13+ 0 5 0 5 5M
E * 11+ 13+ 2 5$ 0 3 3M
O 14 11+ 12- 13+
This can be parsed to create 4 different GFA2 Objects:
- GFA2<BString, OptionalFields>, this create an Object containing all the fields parsed as BString with all the optional tags.\
let parser: GFA2Parser<BString, OptionalFields> = GFA2Parser::new();
let gfa2: GFA2<BString, OptionalFields> = parser
.parse_file(&"./tests/gfa2_files/spec_q7.gfa")
.unwrap();
println!("{:#?}", gfa2);The Object obtained will look like this:
GFA2 {
headers: [
Header {
version: Some("VN:Z:2.0"),
tag: [],
},
Header {
version: Some("ul:Z:https://github.com/pmelsted/GFA-spec/issues7#issuecomment-219685552"),
tag: [],
},
],
segments: [
Segment {
id: "11",
len: "5",
sequence: "ACCTT",
tag: [
OptField {
tag: [112,103],
value: J("pg:J:{\"Human\":[{\"target\":\"chr1\",\"pos\":1500,\"strand\":true}],\"Mouse\":[{\"target\":\"chr2\",\"pos\":2000,\"strand\":false}],\"ecoli\":[{\"target\":\"chr1\",\"pos\":2000,\"strand\":false},{\"target\":\"chr1\",\"pos\":3000,\"strand\":true}]}"),
},
],
},
Segment {
id: "12",
len: "6",
sequence: "TCAAGG",
tag: [],
},
Segment {
id: "13",
len: "7",
sequence: "CTTGATT",
tag: [],
},
],
fragments: [],
edges: [
Edge {
id: "*",
sid1: "11+",
sid2: "12-",
beg1: "1",
end1: "5$",
beg2: "2",
end2: "6$",
alignment: "4M",
tag: [],
},
Edge {
id: "*",
sid1: "12-",
sid2: "13+",
beg1: "0",
end1: "5",
beg2: "0",
end2: "5",
alignment: "5M",
tag: [],
},
Edge {
id: "*",
sid1: "11+",
sid2: "13+",
beg1: "2",
end1: "5$",
beg2: "0",
end2: "3",
alignment: "3M",
tag: [],
},
],
gaps: [],
groups_o: [
GroupO {
id: "14",
var_field: "11+ 12- 13+",
tag: [],
_segment_names: PhantomData,
},
],
groups_u: [],
}
- GFA2<usize, OptionalFields>, this create an Object containing all the fields parsed as usize with all the optional tags.
This kind of conversion use the ASCII codes to convert letters and printable symbols into numbers.
Moreover, this types needs to perform an additional conversion for the Orientation fields [+-] associated with the ref tag of Fragment, Edge, Gap and O-Group, converting the [+] symbol as [0] and [-] as [1].\
let parser: GFA2Parser<usize, OptionalFields> = GFA2Parser::new();
let gfa2: GFA2<usize, OptionalFields> = parser
.parse_file(&"./tests/gfa2_files/spec_q7.gfa")
.unwrap();
println!("{:#?}", gfa2);The Object obtained will look like this:
GFA2 {
headers: [
Header {
version: Some("VN:Z:2.0"),
tag: [],
},
Header {
version: Some("ul:Z:https://github.com/pmelsted/GFA-spec/issues/7#issuecomment-219685552"),
tag: [],
},
],
segments: [
Segment {
id: 11,
len: "5",
sequence: "ACCTT",
tag: [
OptField {
tag: [112,103],
value: J("pg:J:{\"Human\":[{\"target\":\"chr1\",\"pos\":1500,\"strand\":true}],\"Mouse\":[{\"target\":\"chr2\",\"pos\":2000,\"strand\":false}],\"ecoli\":[{\"target\":\"chr1\",\"pos\":2000,\"strand\":false},{\"target\":\"chr1\",\"pos\":3000,\"strand\":true}]}"),
},
],
},
Segment {
id: 12,
len: "6",
sequence: "TCAAGG",
tag: [],
},
Segment {
id: 13,
len: "7",
sequence: "CTTGATT",
tag: [],
},
],
fragments: [],
edges: [
Edge {
id: 42,
sid1: 110,
sid2: 121,
beg1: "1",
end1: "5$",
beg2: "2",
end2: "6$",
alignment: "4M",
tag: [],
},
Edge {
id: 42,
sid1: 121,
sid2: 130,
beg1: "0",
end1: "5",
beg2: "0",
end2: "5",
alignment: "5M",
tag: [],
},
Edge {
id: 42,
sid1: 110,
sid2: 130,
beg1: "2",
end1: "5$",
beg2: "0",
end2: "3",
alignment: "3M",
tag: [],
},
],
gaps: [],
groups_o: [
GroupO {
id: "14",
var_field: "11+ 12- 13+",
tag: [],
_segment_names: PhantomData,
},
],
groups_u: [],
}
In this example the character "*" it's converted as the number "42", that is the DECIMAL code associated with to the character in the ASCII TABLE.
- GFA2<BString, ()>, this create an Object containing all the fields parsed as BString without the optional tags.
- GFA2<usize, ()>, this create an Object containing all the fields parsed as usize without the optional tags.
- A GFA Object can be pretty printed thanks to the implementation of the
Display traits:
let parser: GFA2Parser<BString, OptionalFields> = GFA2Parser::new();
let gfa2: GFA2<BString, OptionalFields> = parser
.parse_file(&"./tests/gfa2_files/spec_q7.gfa")
.unwrap();
println!("{}", gfa2);Obtaining:
H VN:Z:2.0
H ul:Z:https://github.com/pmelsted/GFA-spec/issues/7#issuecomment-219685552
S 11 5 ACCTT pg:J:{"Human":[{"target":"chr1","pos":1500,"strand":true}],"Mouse":[{"target":"chr2","pos":2000,"strand":false}],"ecoli":[{"target":"chr1","pos":2000,"strand":false},{"target":"chr1","pos":3000,"strand":true}]}
S 12 6 TCAAGG
S 13 7 CTTGATT
E * 11+ 12- 1 5$ 2 6$ 4M
E * 12- 13+ 0 5 0 5 5M
E * 11+ 13+ 2 5$ 0 3 3M
O 14 11+ 12- 13+