pandora_lib_promethion/src/annotation/dbsnp.rs

use std::{fmt, str::FromStr};

use bitcode::{Decode, Encode};
use serde::{Deserialize, Serialize};

#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Encode, Decode)]
pub struct DbSnpFreqEntry {
    pub source: String,
    pub values: Vec<Option<f64>>,
}

impl fmt::Display for DbSnpFreqEntry {
    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
        let values = self
            .values
            .iter()
            .map(|v| match v {
                Some(v) => v.to_string(),
                None => ".".to_string(),
            })
            .collect::<Vec<_>>()
            .join(",");

        write!(f, "{}:{}", self.source, values)
    }
}

#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Encode, Decode)]
pub struct DbSnpFreq(pub Vec<DbSnpFreqEntry>);

impl FromStr for DbSnpFreq {
    type Err = anyhow::Error;

    fn from_str(s: &str) -> anyhow::Result<Self> {
        let entries = s
            .split('|')
            .filter(|x| !x.is_empty())
            .map(|entry| {
                let (source, values) = entry
                    .split_once(':')
                    .ok_or_else(|| anyhow::anyhow!("Invalid FREQ entry: {entry}"))?;

                let values = values
                    .split(',')
                    .map(|v| {
                        if v == "." {
                            Ok(None)
                        } else {
                            v.parse::<f64>().map(Some).map_err(|e| {
                                anyhow::anyhow!("Invalid FREQ value `{v}` in {source}: {e}")
                            })
                        }
                    })
                    .collect::<anyhow::Result<Vec<_>>>()?;

                Ok(DbSnpFreqEntry {
                    source: source.to_string(),
                    values,
                })
            })
            .collect::<anyhow::Result<Vec<_>>>()?;

        Ok(DbSnpFreq(entries))
    }
}

impl fmt::Display for DbSnpFreq {
    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
        let s = self
            .0
            .iter()
            .map(|e| e.to_string())
            .collect::<Vec<_>>()
            .join("|");

        write!(f, "{s}")
    }
}

impl DbSnpFreq {
    /// Average ALT frequency across real population sources in a dbSNP FREQ field.
    pub fn maf(&self) -> Option<f32> {
        const EXCLUDED: &[&str] = &["SGDP_PRJ", "dbGaP_PopFreq"];

        let (sum, count) = self
            .0
            .iter()
            .filter_map(|entry| {
                if EXCLUDED.contains(&entry.source.as_str()) {
                    return None;
                }

                let alt = entry.values.get(1).copied().flatten()? as f32;

                if alt <= 0.0 {
                    None
                } else {
                    Some(alt)
                }
            })
            .fold((0.0_f32, 0usize), |(s, c), af| (s + af, c + 1));

        (count > 0).then_some(sum / count as f32)
    }
}