Home

Landing page for documentation that may be updated between releases.

Input Formats

• Guide Library Format

Output Formats

• Count file
• Statistics file

Dual-only

• Query class file

Alignment process

Definitions:

• Query: experimental sequence, e.g. fastq reads
• Targets: designed target sequences, e.g. crispr guide library

High level flow:

1. The targets are reduced to a unique set for each element of a guide set (single, dual).
2. Each query is mapped vs the full set of targets as described here.
3. The best target hit (above defined minimum score) is used to identify the possible guide sets and classification.

Single guide

In the case of single guide the best mapping is linked back to the guide details, in some cases multiple guides may be comparable (see unique_guide). Unless these are exact replicas the mapping becomes a multi-hit, see Read count fields.

Dual guide

Each query of the pair is mapped independently, retaining the mappings with the best score (possible to have a set of mappings with same score).

The mapping set of each query pair is compared vs the guide pairs to identify any matching sets.

Detailing this in prose or pseudo-code is not really helpful, the actual code is more appropriate. This function only deals with the classification process: pycroquet.dualguide.classify_read_pair (v1.5.1)