Crate mzcv 

mzcv

Handle ontologies/controlled vocabularies with getting data from many sources:

1. Download from a URL (only with feature http) [CVIndex::update_from_url]
2. Parse from a given file CVIndex::update_from_path
3. Open binary cache from the standardised cache location
4. Update with values directly in memory CVIndex::update
5. Use statically included data CVSource::static_data

When downloading a file it downloads it to the standardised location and compresses it with gzip compression to not take up too much space. When opening a file (such as a downloaded file) it first parses the file. If it succeeds it places the file at the standardised location and stores the parsed data in the binary cache. If it fails to parse the file it will report the errors both to the caller of the method and leave the errors next to the standard file location for end user convenience.

Lookup

There are three major ways of looking up data: index, name, and fuzzy match search. The first two use HashMaps to do constant time lookups, the second uses a trigram index (when search-index is turned on, or loops over all data if not) and loops over all matches using Levenshtein distance to find good enough matches.

Compilation features

• http allow downloading ontologies from the internet
• serde allow using serde (de)serialise to store data in the cache
• search-index builds a trigram index to speed up fuzzy matching

Macros

accession_code

Create an accession code, using this with a numeric code is match arm valid, using this with an alphanumeric code is const compatible.

curie

A curie. For documentation reasons a full term definition is also allowed. So both examples give the exact same Curie result:

term

Create a new term term!(MS:1002357|PSM-level probability). The accession/name combination is not validated.

Structs

CVFile

The description of a file that is used to built a controlled vocabulary.

CVIndex

An index into a CV which contains the main ways of handling CVs.

CVVersion

Version information for a CV

Curie

A CURIE is a namespace + accession identifier

HashBufReader

A std::io::BufReader inspired design that also calculates the Hash of the read file.

Lines

An iterator over the lines of an instance of HashBufReader.

OboIdentifier

A (usually) unique identifier for an entry in an ontology

OboOntology

An Obo ontology. This can be read from a file with Self::from_file or from a raw reader with Self::from_raw.

OboStanza

An Obo stanza.

OboSynonym

A synonym in an Obo stanza

Term

A term, a CURIE plus its name

Enums

AccessionCode

An accession code, Can either be a numeric code (u32 to 4 milion, so 9 fully utilised digits). Or it can be an ASCII alphanumeric code (case-sensitive) of 1 to 8 characters.

AccessionCodeParseError

An error when parsing an accession code

CURIEParsingError

An error that occured when parsing a CURIE

CVCompression

The used compression of the source CV.

CVError

An error encountered while parsing a CV

ControlledVocabulary

Controlled vocabularies all ontobee listed controlled vocabulaires are present as well as PRIDE and RESID

OboError

All possible errors when reading an Obo file

OboStanzaType

The type for an Obo stanza

OboValue

The value that a property value tag can have in Obo

SynonymScope

The type or scope for a synonym

Traits

CVData

A data element from a CV. Note that technically an implementation could be made that does not have any index, name, or keywords for a data element. Such an element would be kept and stored but would not be accessible via anything else then crate::CVIndex::data.

CVSource

Implement this trait to create a new CV. The best way of using this is with a ZST (zero sized type).

CVStructure

A structure to contain CVData elements but leave the implementation up to the needs of the specific CV.