Added basic molecular formula searching

Author: douweschulte

Cargo.lock

@@ -15,14 +15,14 @@ path = "src/main.rs"
 
 [dependencies]
 clap = { version = "4.5", features = ["derive", "cargo"] }
-colored = "2"
-itertools = "0.13"
+colored = "3"
+itertools = "0.14"
 rayon = "1.10"
 rustyms = "0.9.0"
 
 [patch.crates-io]
 # rustyms = { git = "https://github.com/snijderlab/rustyms", branch = "main" }
-#rustyms = { path = "../rustyms/rustyms" }
+rustyms = { path = "../rustyms/rustyms" }
 
 [profile.release]
 debug = true

Cargo.toml

@@ -170,6 +170,11 @@ pub struct Cli {
     /// Multiple positions can be specified by using this argument multiple times.
     #[arg(long, value_parser=positions_parser)]
     pub positions: Option<Vec<(Vec<AminoAcid>, Position)>>,
+
+    /// Search for a fitting molecular formula for this mass.
+    // Contains the mass and number of digits
+    #[arg(long = "formula", value_parser=formula_parser)]
+    pub formula_target: Option<(Mass, usize)>,
 }
 
 impl Cli {
@@ -189,6 +194,23 @@ impl Cli {
     }
 }
 
+fn formula_parser(value: &str) -> Result<(Mass, usize), String> {
+    let target = Mass::new::<rustyms::system::dalton>(value.parse::<f64>().map_err(|err| {
+        format!("Given target mass for formula search is not a valid number: {err}")
+    })?);
+    Ok((
+        target,
+        if let Some((_, tail)) = value.split_once('.') {
+            tail.to_lowercase()
+                .split_once('e')
+                .map_or(tail, |(t, _)| t)
+                .len()
+        } else {
+            0
+        },
+    ))
+}
+
 fn positions_parser(value: &str) -> Result<(Vec<AminoAcid>, Position), String> {
     value
         .split_once('@')

src/cli.rs

@@ -5,7 +5,7 @@ use itertools::Itertools;
 use rayon::prelude::*;
 use rustyms::align::par_consecutive_align;
 use rustyms::imgt::Selection;
-use rustyms::system::Mass;
+use rustyms::system::{dalton, Mass};
 use rustyms::{
     align::*,
     find_isobaric_sets, imgt,
@@ -18,6 +18,8 @@ use rustyms::{
     AminoAcid, AtMax, Chemical, MassMode, MolecularFormula, Multi, Peptidoform, SimpleLinear,
     Tolerance, UnAmbiguous,
 };
+use rustyms::{find_formulas, Element};
+use std::num::NonZeroU16;
 use std::{
     collections::HashSet,
     io::{BufWriter, Write},
@@ -414,6 +416,32 @@ fn main() {
             }
             display_germline(allele, &args);
         }
+    } else if let Some(target) = args.formula_target {
+        const DEFAULT_ELEMENTS: &[(Element, Option<NonZeroU16>)] = &[
+            (Element::H, None),
+            (Element::C, None),
+            (Element::O, None),
+            (Element::N, None),
+            (Element::S, None),
+        ];
+        let mut data = vec![["Formula".to_string(), "Mass".to_string()]];
+        for formula in find_formulas(
+            target.0,
+            Tolerance::Absolute(Mass::new::<dalton>(10.0_f64.powf(-(target.1 as f64)) / 2.0)),
+            DEFAULT_ELEMENTS,
+        )
+        .iter()
+        {
+            data.push([
+                formula.hill_notation_fancy(),
+                display_mass(
+                    formula.monoisotopic_mass(),
+                    false,
+                    Some((target.1 + 2).max(3)),
+                ),
+            ]);
+        }
+        table(&data, true, &[Styling::none(), Styling::none()]);
     } else {
         println!("Please provide an argument to work with, use --help to see all options.")
     }