Struct SimpleErrorUmiAssigner 

pub struct SimpleErrorUmiAssigner { /* private fields */ }

Simple error-tolerant UMI assigner

Groups UMIs that are within a specified edit distance using a simple clustering algorithm. When a new UMI is encountered, it is compared against existing groups. If it matches any UMI in a group, it is added to that group. If it matches multiple groups, those groups are merged. This can lead to transitive clustering where UMIs are grouped together even if they differ by more than the threshold.

Algorithm

1. For each UMI, find all existing groups that contain at least one UMI within threshold
2. If no matches, create a new group
3. If one match, add to that group
4. If multiple matches, merge all matching groups

Thread Safety

Uses atomic counter for ID generation, making it safe to use across threads.

Implementations

impl SimpleErrorUmiAssigner

pub fn new(max_mismatches: u32) -> SimpleErrorUmiAssigner

Create a new simple error-tolerant UMI assigner

Arguments

• max_mismatches - Maximum number of mismatches allowed between UMIs in the same group

Returns

A new SimpleErrorUmiAssigner instance

Trait Implementations

impl UmiAssigner for SimpleErrorUmiAssigner

fn assign(&self, raw_umis: &[String]) -> Vec<MoleculeId>

Assign molecule IDs to UMIs

fn as_any(&self) -> &(dyn Any + 'static)

Downcast to concrete type (for pattern matching)

fn is_same_umi(&self, a: &str, b: &str) -> bool

Check if two UMIs are the same

fn canonicalize(&self, umi: &str) -> String

Get canonical form of UMI

fn split_templates_by_pair_orientation(&self) -> bool

Whether to split templates by pair orientation