adding seqspec rust tooling for python-rust parity

Author: sbooeshaghi

Cargo.toml

@@ -13,12 +13,15 @@ crate-type = ["rlib", "cdylib"]
 
 [dependencies]
 clap = { version = "4.5.46", features = ["derive"] }
+jsonschema = "0.33.0"
 pyo3 = { version = "0.25", optional = true, features = ["extension-module", "abi3-py312"] }
 # pythonize = "0.25.0"
 serde = { version = "1", features = ["derive"] }
 serde_json = "1"
 serde_yaml = "0.9"
 thiserror = "1"
+reqwest = { version = "0.12", features = ["blocking", "rustls-tls"] }
+flate2 = "1"
 
 
 [features]

README.md

@@ -37,3 +37,22 @@ Ali Sina Booeshaghi, Xi Chen, Lior Pachter, A machine-readable specification for
 - [Contribute a `seqspec` : `docs/CONTRIBUTING.md`](docs/CONTRIBUTING.md)
 - [Watch a YouTube video about `seqspec`](https://youtu.be/NSj6Vpzy8tU)
 - [Read the manuscript that describes `seqspec`](https://doi.org/10.1093/bioinformatics/btae168)
+
+## Rust implementation
+
+- [] build : Generate a complete seqspec with natural language.
+- [x] check : Validate seqspec file against specification (verify check)
+- [x] find : Find objects in seqspec file
+- [x] file : List files present in seqspec file
+- [x] format : Autoformat seqspec file
+- [x] index : Identify position of elements in seqspec file
+- [x] info : Get information from seqspec file
+- [x] init : Generate a new empty seqspec file
+- [x] insert : Insert regions or reads into an existing spec (TODO: move Input structs to models)
+- [x] methods : Convert seqspec file into methods section
+- [x] modify : Modify attributes of various elements in seqspec file
+- [x] onlist : Get onlist file for elements in seqspec file
+- [] print : Display the sequence and/or library structure from seqspec file
+- [x] split : Split seqspec file by modality
+- [x] upgrade : Upgrade seqspec file to current version
+- [x] version: Get seqspec tool version and seqspec file version

seqspec/seqspec_index.py

@@ -353,9 +353,11 @@ def get_coordinate_by_read_id(spec: Assay, modality: str, read_id: str) -> Coord
 
     return coord
 
+
 FEATURE_REGION_TYPES = {"CDNA", "GDNA", "PROTEIN", "TAG", "SGRNA_TARGET"}
 
-def format_kallisto_bus(indices: List[Coordinate], subregion_type=None):
+
+def format_kallisto_bus(indices: List[Coordinate], subregion_type=None) -> str:
     bcs = []
     umi = []
     feature = []
@@ -376,7 +378,9 @@ def format_kallisto_bus(indices: List[Coordinate], subregion_type=None):
     return x
 
 
-def format_kallisto_bus_force_single(indices: List[Coordinate], subregion_type=None):
+def format_kallisto_bus_force_single(
+    indices: List[Coordinate], subregion_type=None
+) -> str:
     bcs = []
     umi = []
     feature = []
@@ -408,7 +412,7 @@ def format_kallisto_bus_force_single(indices: List[Coordinate], subregion_type=N
 
 # this one should only return one string
 # TODO: return to this
-def format_seqkit_subseq(indices: List[Coordinate], subregion_type=None):
+def format_seqkit_subseq(indices: List[Coordinate], subregion_type=None) -> str:
     # The x string format is start:stop (1-indexed)
     # x = ""
     # region = indices[0]
@@ -422,7 +426,7 @@ def format_seqkit_subseq(indices: List[Coordinate], subregion_type=None):
     return x
 
 
-def format_tab(indices: List[Coordinate], subregion_type=None):
+def format_tab(indices: List[Coordinate], subregion_type=None) -> str:
     x = ""
     for idx, coord in enumerate(indices):
         rcv = coord.rcv
@@ -433,7 +437,7 @@ def format_tab(indices: List[Coordinate], subregion_type=None):
     return x[:-1]
 
 
-def format_starsolo(indices: List[Coordinate], subregion_type=None):
+def format_starsolo(indices: List[Coordinate], subregion_type=None) -> str:
     bcs = []
     umi = []
     cdna = []
@@ -451,7 +455,7 @@ def format_starsolo(indices: List[Coordinate], subregion_type=None):
     return x
 
 
-def format_simpleaf(indices: List[Coordinate], subregion_type=None):
+def format_simpleaf(indices: List[Coordinate], subregion_type=None) -> str:
     x = ""
     xl = []
     for idx, coord in enumerate(indices):
@@ -469,7 +473,7 @@ def format_simpleaf(indices: List[Coordinate], subregion_type=None):
     return "".join(xl)
 
 
-def format_zumis(indices: List[Coordinate], subregion_type=None):
+def format_zumis(indices: List[Coordinate], subregion_type=None) -> str:
     xl = []
     for idx, coord in enumerate(indices):
         x = ""
@@ -486,7 +490,7 @@ def format_zumis(indices: List[Coordinate], subregion_type=None):
 
 
 def stable_deduplicate_fqs(fqs):
-    # stably deduplicate gdna_fqs
+    # stably deduplicate fqs
     seen_fqs = set()
     deduplicated_fqs = []
     for r in fqs:
@@ -496,7 +500,7 @@ def stable_deduplicate_fqs(fqs):
     return deduplicated_fqs
 
 
-def format_chromap(indices: List[Coordinate], subregion_type=None):
+def format_chromap(indices: List[Coordinate], subregion_type=None) -> str:
     bc_fqs = []
     bc_str = []
     gdna_fqs = []
@@ -563,7 +567,7 @@ def filter_groupby_region_type(g, keep=["umi", "barcode", "cdna"]):
     return g
 
 
-def format_relative(indices: List[Coordinate], subregion_type=None):
+def format_relative(indices: List[Coordinate], subregion_type=None) -> str:
     x = ""
     for idx, coord in enumerate(indices):
         rg_strand = coord.strand  # noqa
@@ -622,9 +626,6 @@ def groupby_region_type(rgns):
     return d
 
 
-# def group_regions_by_region_type(rgns):
-
-
 def format_splitcode_row(obj, rgncdiffs, idx=0, rev=False, complement=False):
     # print(obj.region_id, idx)
     # TODO only have one object left and one object right of the sequence
@@ -689,7 +690,7 @@ def format_splitcode_row(obj, rgncdiffs, idx=0, rev=False, complement=False):
     return {"region_type": obj.region_type, "fmt": e}
 
 
-def format_splitcode(indices: List[Coordinate], subregion_type=None):
+def format_splitcode(indices: List[Coordinate], subregion_type=None) -> str:
     # extraction based on fixed sequences
     # extraction based on onlist sequences
     # umi - bc3 - link2 - bc2 - link1 - bc1 - read

src/lib.rs

@@ -4,6 +4,18 @@ pub use models::{file, region, read, onlist, assay};
 pub mod utils;
 pub mod seqspec_version;
 pub mod seqspec_format;
+pub mod seqspec_find;
+pub mod seqspec_index;
+pub mod seqspec_file;
+pub mod seqspec_split;
+pub mod seqspec_info;
+pub mod seqspec_init;
+pub mod seqspec_methods;
+pub mod seqspec_modify;
+pub mod seqspec_upgrade;
+pub mod seqspec_insert;
+pub mod seqspec_check;
+pub mod seqspec_onlist;
 
 // #[cfg(feature = "python-binding")]
 // mod py_module;  // lives in src/py_module.rs

src/main.rs

@@ -5,6 +5,18 @@
 
 use seqspec::seqspec_version;
 use seqspec::seqspec_format;
+use seqspec::seqspec_find;
+use seqspec::seqspec_index;
+use seqspec::seqspec_file;
+use seqspec::seqspec_split;
+use seqspec::seqspec_info;
+use seqspec::seqspec_init;
+use seqspec::seqspec_methods;
+use seqspec::seqspec_modify;
+use seqspec::seqspec_upgrade;
+use seqspec::seqspec_insert;
+use seqspec::seqspec_check;
+use seqspec::seqspec_onlist;
 use seqspec::utils;
 
 use clap::{Parser, Subcommand};
@@ -20,13 +32,37 @@ struct Args {
 enum Commands {
     Version(seqspec_version::VersionArgs),
     Format(seqspec_format::FormatArgs),
+    Find(seqspec_find::FindArgs),
+    Index(seqspec_index::IndexArgs),
+    File(seqspec_file::FileArgs),
+    Split(seqspec_split::SplitArgs),
+    Info(seqspec_info::InfoArgs),
+    Init(seqspec_init::InitArgs),
+    Methods(seqspec_methods::MethodsArgs),
+    Modify(seqspec_modify::ModifyArgs),
+    Upgrade(seqspec_upgrade::UpgradeArgs),
+    Insert(seqspec_insert::InsertArgs),
+    Check(seqspec_check::CheckArgs),
+    Onlist(seqspec_onlist::OnlistArgs),
     // other subcommands later...
 }
 
 fn main() {
     let args = Args::parse();
     match args.subcmd {
-        Commands::Version(args) => seqspec_version::validate_version_args(&args),
-        Commands::Format(args) => seqspec_format::validate_format_args(&args),
+        Commands::Version(args) => seqspec_version::run_version(&args),
+        Commands::Format(args) => seqspec_format::run_format(&args),
+        Commands::Find(args) => seqspec_find::run_find(&args),
+        Commands::Index(args) => seqspec_index::run_index(&args),
+        Commands::File(args) => seqspec_file::run_file(&args)
+        ,Commands::Split(args) => seqspec_split::run_split(&args)
+        ,Commands::Info(args) => seqspec_info::run_info(&args)
+        ,Commands::Init(args) => seqspec_init::run_init(&args)
+        ,Commands::Methods(args) => seqspec_methods::run_methods(&args)
+        ,Commands::Modify(args) => seqspec_modify::run_modify(&args)
+        ,Commands::Upgrade(args) => seqspec_upgrade::run_upgrade(&args)
+        ,Commands::Insert(args) => seqspec_insert::run_insert(&args)
+        ,Commands::Check(args) => { seqspec_check::run_check(&args); }
+        ,Commands::Onlist(args) => seqspec_onlist::run_onlist(&args)
     }
 }

src/models/assay.rs

@@ -52,6 +52,9 @@ pub struct Assay {
     pub library_spec:  Vec<Region>,
 }
 
+pub enum Codec { Yaml, Json }
+
+
 impl Assay {
     pub fn new(
         assay_id: String,
@@ -85,8 +88,21 @@ impl Assay {
         serde_json::to_string(self)
     }
 
+    pub fn to_bytes(&self, codec: Codec) -> Result<Vec<u8>, Box<dyn std::error::Error>> {
+        Ok(match codec {
+            Codec::Yaml => serde_yaml::to_string(self)?.into_bytes(),
+            Codec::Json => serde_json::to_vec(self)?,
+        })
+    }
+    pub fn from_bytes(bytes: &[u8], codec: Codec) -> Result<Self, Box<dyn std::error::Error>> {
+        Ok(match codec {
+            Codec::Yaml => serde_yaml::from_slice(bytes)?,
+            Codec::Json => serde_json::from_slice(bytes)?,
+        })
+    }
+
     // Core helpers ----------------------------------------------------
-    pub fn update_spec(&mut self) {
+    pub fn update_spec(&mut self) -> () {
         for r in &mut self.library_spec {
             r.update_attr();
         }
@@ -110,9 +126,9 @@ impl Assay {
 
     pub fn get_read(&self, read_id: &str) -> Option<Read> {
         self.sequence_spec
-            .iter()
-            .find(|r| r.read_id == read_id)
-            .cloned()
+        .iter()
+        .find(|r| r.read_id == read_id)
+        .cloned()
     }
 
     /// Insert regions under the top-level region for `modality`.

src/models/coordinate.rs

@@ -0,0 +1,17 @@
+use serde::{Serialize, Deserialize};
+
+use crate::models::region::RegionCoordinate;
+
+#[derive(Debug, Serialize, Deserialize)]
+pub struct Coordinate {
+    pub query_id: String,
+    pub query_name: String,
+    pub query_type: String,
+    pub rcv: Vec<RegionCoordinate>,
+    #[serde(default = "default_strand")]
+    pub strand: String,
+}
+
+fn default_strand() -> String {
+    "pos".to_string()
+}

src/models/mod.rs

@@ -2,4 +2,5 @@ pub mod assay;
 pub mod file;
 pub mod onlist;
 pub mod read;
-pub mod region;
+pub mod region;
+pub mod coordinate;

src/models/region.rs

@@ -156,7 +156,7 @@ impl Region {
         use std::collections::BTreeSet;
         let mut set = BTreeSet::new();
         for r in self.get_leaves() {
-            set.insert(r.region_type);
+            set.insert(r.region_type.clone());
         }
         set.into_iter().collect()
     }