feat: transform "db to-bin" to "strucvars txt-to-bin" (#218) (#219)

README.md

@@ -14,15 +14,15 @@ They are written in the Rust programming language to speed up the execution of c
 At the moment, the following sub commands exist:
 
 - `db` -- subcommands to build binary (protobuf) database files
-    - `db to-bin` -- convert text files downloaded by [varfish-db-downloader](https://github.com/bihealth/varfish-db-downloader/) to binary for fast use in query sub commands
-    - `db mk-inhouse` -- compile per-case structural variant into an in-house database previously created by `db compile`
 - `seqvars` -- subcommands for processing sequence (aka small/SNV/indel) variants
     - `seqvars ingest` -- convert single VCF file into internal format for use with `seqvars query`
+    - `seqvars query` -- perform sequence variant filtration and on-the-fly annotation
     - `seqvars prefilter` -- limit the result of `seqvars prefilter` by population frequency and/or distance to exon
     - `seqvars aggregate` -- read through multiple VCF files written by `seqvars ingest` and computes a carrier counts table.
-    - `seqvars query` -- perform sequence variant filtration and on-the-fly annotation
 - `strucvars` -- subcommands for processing structural (aka large variants, CNVs, etc.) variants
     - `strucvars ingest` -- convert one or more structural variant files for use with `strucvars query`
+    - `strucvars aggregate` -- compile per-case structural variant into an in-house database, to be converted to `.bin` with `strucvars txt-to-bin`.
+    - `strucvars txt-to-bin` -- convert text files downloaded by [varfish-db-downloader](https://github.com/bihealth/varfish-db-downloader/) to binary for fast use in `strucvars query` commands
     - `strucvars query` -- perform structural variant filtration and on-the-fly annotation
 
 ## Overall Design
@@ -39,40 +39,6 @@ The worker will create a result file that can be directly imported by the server
 
 Future versions may provide persistently running HTTP/REST servers that provide functionality without startup cost.
 
-## The `db to-bin` Command
-
-Convert output of [varfish-db-downloader](https://github.com/bihealth/varfish-db-downloader/) to a directory with databases to be used by query commands such as `strucvars query`.
-
-```
-$ varfish-server-worker db to-bin \
-    --input-type {ClinvarSv,StrucvarInhouse,...} \
-    --path-input IN.txt \
-    --path-output-bin DST.bin
-```
-
-## The `db mk-inhouse` Command
-
-Import multiple files created by `strucvars ingest` into a database previously created by `db compile`.
-You can specify the files individually.
-Paths starting with an at (`@`) character are interpreted as files with lists of paths.
-You can mix paths with `@` and without.
-
-```
-$ varfish-server-worker db mk-inhouse \
-    --genome-release {Grch37,Grch38} \
-    --path-output-tsv OUT.tsv \
-    --path-input-tsvs IN/file1.gts.tsv.gz \
-    [--path-input-tsv IN/file1.gts.tsv.gz] \
-
-# OR:
-
-$ varfish-server-worker db mk-inhouse \
-    --genome-release {Grch37,Grch38} \
-    --path-output-tsv OUT.tsv \
-    --path-input-tsvs @IN/path-list.txt \
-    [--path-input-tsvs @IN/path-list2.txt]
-```
-
 ## The `seqvars ingest` Command
 
 This command takes as the input a single VCF file from a (supported) variant caller and converts it into a file for further querying.
@@ -286,6 +252,40 @@ Overall, the command will emit the following header rows in addition to the `##c
 > It only merges the input VCF files from multiple callers (all files must have the same samples) and converts them into the internal format.
 > The `INFO/annsv` field is filled by `strucvars query`.
 
+## The `strucvars aggregate` Command
+
+Import multiple files created by `strucvars ingest` into a database that can be convered to `.bin` with `strucvars txt-to-bin` and then used by `strucvars query`.
+You can specify the files individually.
+Paths starting with an at (`@`) character are interpreted as files with lists of paths.
+You can mix paths with `@` and without.
+
+```
+$ varfish-server-worker strucvars aggregate \
+    --genome-release {Grch37,Grch38} \
+    --path-output OUT.tsv \
+    --path-input IN/file1.vcf.gz \
+    [--path-input IN/file1.vcf.gz] \
+
+# OR:
+
+$ varfish-server-worker db mk-inhouse \
+    --genome-release {Grch37,Grch38} \
+    --path-output OUT.tsv \
+    --path-input @IN/path-list.txt \
+    [--path-input @IN/path-list2.txt]
+```
+
+## The `strucvars txt-to-bin` Command
+
+Convert output of [varfish-db-downloader](https://github.com/bihealth/varfish-db-downloader/) to a directory with databases to be used by query commands such as `strucvars query`.
+
+```
+$ varfish-server-worker strucvars txt-to-bin \
+    --input-type {ClinvarSv,StrucvarInhouse,...} \
+    --path-input IN.txt \
+    --path-output-bin DST.bin
+```
+
 # Developer Information
 
 This section is only relevant for developers of `varfish-server-worker`.

src/common.rs

@@ -511,6 +511,52 @@ where
     Ok(buffer)
 }
 
+/// Enum for the supported gene/transcript databases.
+#[derive(
+    serde::Serialize,
+    serde::Deserialize,
+    enum_map::Enum,
+    PartialEq,
+    Eq,
+    Clone,
+    Copy,
+    Debug,
+    Default,
+    strum::EnumString,
+    strum::Display,
+)]
+#[serde(rename_all = "lowercase")]
+#[strum(serialize_all = "lowercase")]
+pub enum Database {
+    /// RefSeq
+    #[default]
+    RefSeq,
+    /// ENSEMBL
+    Ensembl,
+}
+
+/// Enum for the supported TADs.
+#[derive(
+    serde::Serialize,
+    serde::Deserialize,
+    enum_map::Enum,
+    PartialEq,
+    Eq,
+    Clone,
+    Copy,
+    Debug,
+    Default,
+    strum::EnumString,
+    strum::Display,
+)]
+#[serde(rename_all = "lowercase")]
+#[strum(serialize_all = "lowercase")]
+pub enum TadSet {
+    /// hESC
+    #[default]
+    Hesc,
+}
+
 #[cfg(test)]
 mod test {
     use noodles_vcf as vcf;

src/main.rs

@@ -1,7 +1,6 @@
 //! VarFish Server Worker main executable
 
 pub mod common;
-pub mod db;
 pub mod seqvars;
 pub mod strucvars;
 
@@ -30,30 +29,12 @@ struct Cli {
 #[allow(clippy::large_enum_variant)]
 #[derive(Debug, Subcommand)]
 enum Commands {
-    /// Database building related commands.
-    Db(Db),
     /// Structural variant related commands.
     Strucvars(Strucvars),
     /// Sequence variant related commands.
     Seqvars(Seqvars),
 }
 
-/// Parsing of "db *" sub commands.
-#[derive(Debug, Args)]
-#[command(args_conflicts_with_subcommands = true)]
-struct Db {
-    /// The sub command to run
-    #[command(subcommand)]
-    command: DbCommands,
-}
-
-/// Enum supporting the parsing of "db *" sub commands.
-#[allow(clippy::large_enum_variant)]
-#[derive(Debug, Subcommand)]
-enum DbCommands {
-    ToBin(db::to_bin::cli::Args),
-}
-
 /// Parsing of "sv *" sub commands.
 #[derive(Debug, Args)]
 #[command(args_conflicts_with_subcommands = true)]
@@ -69,6 +50,7 @@ enum StrucvarsCommands {
     Aggregate(strucvars::aggregate::cli::Args),
     Ingest(strucvars::ingest::Args),
     Query(strucvars::query::Args),
+    TxtToBin(strucvars::txt_to_bin::cli::Args),
 }
 
 /// Parsing of "seqvars *" sub commands.
@@ -111,11 +93,6 @@ fn main() -> Result<(), anyhow::Error> {
     let term = Term::stderr();
     tracing::subscriber::with_default(collector, || {
         match &cli.command {
-            Commands::Db(db) => match &db.command {
-                DbCommands::ToBin(args) => {
-                    db::to_bin::cli::run(&cli.common, args)?;
-                }
-            },
             Commands::Seqvars(seqvars) => match &seqvars.command {
                 SeqvarsCommands::Aggregate(args) => {
                     seqvars::aggregate::run(&cli.common, args)?;
@@ -137,6 +114,9 @@ fn main() -> Result<(), anyhow::Error> {
                 StrucvarsCommands::Query(args) => {
                     strucvars::query::run(&cli.common, args)?;
                 }
+                StrucvarsCommands::TxtToBin(args) => {
+                    strucvars::txt_to_bin::cli::run(&cli.common, args)?;
+                }
             },
         }
 

src/strucvars/mod.rs

@@ -1,3 +1,5 @@
 pub mod aggregate;
 pub mod ingest;
+pub mod pbs;
 pub mod query;
+pub mod txt_to_bin;

src/strucvars/query/bgdbs.rs

@@ -10,8 +10,8 @@ use strum_macros::{Display, EnumString};
 use tracing::info;
 
 use crate::{
-    common::{trace_rss_now, CHROMS},
-    db::{conf::GenomeRelease, pbs},
+    common::{trace_rss_now, GenomeRelease, CHROMS},
+    strucvars::pbs,
 };
 
 use super::{

src/strucvars/query/clinvar.rs

@@ -6,10 +6,7 @@ use prost::Message;
 use thousands::Separable;
 use tracing::{info, warn};
 
-use crate::{
-    common::{reciprocal_overlap, CHROMS},
-    db::conf::GenomeRelease,
-};
+use crate::common::{reciprocal_overlap, GenomeRelease, CHROMS};
 
 use super::{
     records::ChromRange,