# Mappers: "Why do we use that mapper? I dunno 🤷‍♂️️ because pop-gen do it?"

### Background

- A lot of inherited knowledge from 'older' ancient DNA fields, or just the 'tradition' within the lab
- e.g. everyone in our group uses 'bwa'... why? 🤷‍♀️
- We _assume_ it's best, but if we ask, no-one can explain why
- Both for settings and which tools:
- all existing studies focused on human poop-gen or animals: Oliva 2021, Poullet 2020
- only test 'established' tools (`bowtie2`/`bwa`)

### Open Questions

- Which tool to use for microbial genomes?
- What settings to use?
- When dealing with community-diverse samples (e.g. dealing with close/distant relatives)
- How much effect does relaxing mismatches cause for bringing in environmental relatives
- Are more modern mappers suitable for aDNA (e.g. `minimap2`)
- Do we need 'damage aware' dedicated aDNA mappers (e.g. `mapAD`)

### Project idea

Benchmarking of established and newer aligners with different amount of damage/length for aligning against microbial genomes

Variables:

- Mapper (`bowtie2`/`bwa aln`/`bwa mem`/`minimap2`/`mapAD`)
- Microbe/Genome (bacteria/virus - with close relatives)
- Damage/Length (spanning typical aDNA ranges)
- Mapper's Parameters

### References

Poullet, M., & Orlando, L. (2020). Assessing DNA Sequence Alignment Methods for Characterizing Ancient Genomes and Methylomes. Frontiers in Ecology and Evolution, 8, 105. https://doi.org/10.3389/fevo.2020.00105

Oliva, A., Tobler, R., Llamas, B., & Souilmi, Y. (2021). Additional evaluations show that specific BWA‐aln settings still outperform BWA‐mem for ancient DNA data alignment. Ecology and Evolution. https://doi.org/10.1002/ece3.8297

---

## Brainstrom

- Everyone uses `bwa aln` (more for eukarotic, microbes sometimes Bowtie2)
- But bowtie2 is faster (multithreads indexing), and accordingly to Poullet and Ludovic
- Minimap2
- Surprisingly OK with aDNA if you look at how much of the ends of reads were clipped?