feat: add confidence threshold option --conf

README.md

@@ -37,7 +37,7 @@ sequencing technology. Read more about the development of this method [here][pap
 
 [![Conda (channel only)](https://img.shields.io/conda/vn/bioconda/nohuman)](https://anaconda.org/bioconda/nohuman)
 [![bioconda version](https://anaconda.org/bioconda/nohuman/badges/platforms.svg)](https://anaconda.org/bioconda/nohuman)
-![Conda](https://img.shields.io/conda/dn/bioconda/nohuman)
+![Conda Downloads](https://img.shields.io/conda/d/bioconda/nohuman)
 
 ```shell
 $ conda install -c bioconda nohuman
@@ -187,6 +187,12 @@ or to specify a different path for the output
 $ nohuman -t 4 --out1 clean_1.fq --out2 clean_2.fq in_1.fq in_2.fq
 ```
 
+Set a [minimum confidence score][conf] for kraken2 classifications
+
+```
+$ nohuman --conf 0.5 in.fq
+```
+
 > [!TIP]
 > Compressed output will be inferred from the specified output path(s). If no output path is provided, the same
 > compression as the input will be used. To override the output compression format, use the `--output-type` option. 
@@ -215,6 +221,7 @@ Options:
   -F, --output-type <FORMAT>  Output compression format. u: uncompressed; b: Bzip2; g: Gzip; x: Xz (Lzma); z: Zstd
   -t, --threads <INT>         Number of threads to use in kraken2 and optional output compression. Cannot be 0 [default: 1]
   -H, --human                 Output human reads instead of removing them
+  -C, --conf <[0, 1]>         Kraken2 minimum confidence score [default: 0.0]
   -v, --verbose               Set the logging level to verbose
   -h, --help                  Print help (see more with '--help')
   -V, --version               Print version
@@ -275,7 +282,12 @@ Options:
 
   -H, --human
           Output human reads instead of removing them
+          
+  -C, --conf <[0, 1]>
+          Kraken2 minimum confidence score
 
+          [default: 0.0]
+          
   -v, --verbose
           Set the logging level to verbose
 
@@ -326,4 +338,6 @@ more details and for other alternate approaches.
 
 [paper]: https://doi.org/10.1093/gigascience/giae010
 
-[ghcr]: https://github.com/mbhall88/nohuman/pkgs/container/nohuman
+[ghcr]: https://github.com/mbhall88/nohuman/pkgs/container/nohuman
+
+[conf]: https://github.com/DerrickWood/kraken2/blob/master/docs/MANUAL.markdown#confidence-scoring

src/lib.rs

@@ -157,6 +157,16 @@ pub fn validate_db_directory(path: &Path) -> Result<PathBuf, String> {
     ))
 }
 
+/// Parse confidence score from the command line. Will be passed on to kraken2. Must be in the
+/// closed interval [0, 1] - i.e. 0 <= confidence <= 1.
+pub fn parse_confidence_score(s: &str) -> Result<f32, String> {
+    let confidence: f32 = s.parse().map_err(|_| "Confidence score must be a number")?;
+    if !(0.0..=1.0).contains(&confidence) {
+        return Err("Confidence score must be in the closed interval [0, 1]".to_string());
+    }
+    Ok(confidence)
+}
+
 #[cfg(test)]
 mod tests {
     use super::*;
@@ -205,4 +215,25 @@ mod tests {
         let expected = PathBuf::from("Cargo.toml");
         assert_eq!(actual, expected)
     }
+
+    #[test]
+    fn test_parse_confidence_score() {
+        let result = parse_confidence_score("0.5");
+        assert!(result.is_ok());
+        assert_eq!(result.unwrap(), 0.5);
+
+        let result = parse_confidence_score("1.0");
+        assert!(result.is_ok());
+        assert_eq!(result.unwrap(), 1.0);
+
+        let result = parse_confidence_score("0.0");
+        assert!(result.is_ok());
+        assert_eq!(result.unwrap(), 0.0);
+
+        let result = parse_confidence_score("1.1");
+        assert!(result.is_err());
+
+        let result = parse_confidence_score("-0.1");
+        assert!(result.is_err());
+    }
 }

src/main.rs

@@ -8,7 +8,8 @@ use env_logger::Builder;
 use log::{debug, error, info, warn, LevelFilter};
 use nohuman::compression::CompressionFormat;
 use nohuman::{
-    check_path_exists, download::download_database, validate_db_directory, CommandRunner,
+    check_path_exists, download::download_database, parse_confidence_score, validate_db_directory,
+    CommandRunner,
 };
 
 static DEFAULT_DB_LOCATION: LazyLock<String> = LazyLock::new(|| {
@@ -72,6 +73,10 @@ struct Args {
     #[arg(short = 'H', long = "human")]
     keep_human_reads: bool,
 
+    /// Kraken2 minimum confidence score
+    #[arg(short = 'C', long = "conf", value_name = "[0, 1]", default_value = "0.0", value_parser = parse_confidence_score)]
+    confidence: f32,
+
     /// Set the logging level to verbose
     #[arg(short, long)]
     verbose: bool,
@@ -142,6 +147,7 @@ fn main() -> Result<()> {
     let temp_kraken_output =
         tempfile::NamedTempFile::new().context("Failed to create temporary kraken output file")?;
     let threads = args.threads.to_string();
+    let confidence = args.confidence.to_string();
     let db = validate_db_directory(&args.database)
         .map_err(|e| anyhow::anyhow!(e))?
         .to_string_lossy()
@@ -153,6 +159,8 @@ fn main() -> Result<()> {
         &db,
         "--output",
         temp_kraken_output.path().to_str().unwrap(),
+        "--confidence",
+        &confidence,
     ];
     match input.len() {
         0 => bail!("No input files provided"),