ALPACA/PANGAIA ANNUAL WORKSHOP - Hackathon

Sequence to graph aligner evaluation

Aligners

vg

Relevant links:

• https://github.com/vgteam/vg
• map: https://www.nature.com/articles/nbt.4227#Abs2
• giraffe: https://www.science.org/doi/full/10.1126/science.abg8871

GraphAligner

Relevant links:

• https://github.com/maickrau/GraphAligner
• https://genomebiology.biomedcentral.com/articles/10.1186/s13059-020-02157-2

minigraph

Relevant links:

• https://github.com/lh3/minigraph
• https://link.springer.com/article/10.1186/s13059-020-02168-z

GraphChainer

Relevant links:

• https://github.com/algbio/GraphChainer
• https://academic.oup.com/bioinformatics/article/39/8/btad460/7231478

minichain

Relevant links:

• https://github.com/at-cg/minichain
• https://link.springer.com/chapter/10.1007/978-3-031-29119-7_4

PanAligner

Relevant links:

• https://github.com/at-cg/PanAligner
• https://www.biorxiv.org/content/10.1101/2023.06.21.545871v1.abstract

PaSGAL

Reasons for investigation:

• Output format is not canonical gaf/gam, investigation is needed to see if a conversion is feasible

Relevant links:

• https://github.com/ParBLiSS/PaSGAL
• https://doi.org/10.1109/IPDPS.2019.00055

V-Align

Reasons for investigation:

• The instructions are not clear
• It may have everything it needs to run it
• It is not clear what are inputs and outputs formats

Relevant links:

• https://github.com/tcsatc/V-ALIGN
• https://www.liebertpub.com/doi/abs/10.1089/cmb.2017.0264

Vargas

Reasons for failure:

• It needs a specific graph format as input, that must be constructed with vargas define
• It is not possible to convert gfa/vg to such format
• It is thus impossible to fairly compare with other methods

Relevant links:

• https://github.com/langmead-lab/vargas
• https://academic.oup.com/bioinformatics/article/36/12/3712/5823884

HGA

Reasons for failure:

• It need its proprietary input format, for which the only "documentation" is in RapidsAtHKUST/hga#1
• It is not possible to convert gfa/vg to such format
• The output format is not specified, and it is currently impossible to run the tool
• It is thus impossible to fairly compare with other methods

Relevant links:

• https://github.com/RapidsAtHKUST/hga
• https://doi.org/10.1145/3472456.3472505

ASTARIX2

• The output format is a TSV and doesn't have a gaf/gam conversion.
• It may be possible to have a conversion tool, but the task is not trivial and the authors are not progressing on it, see eth-sri/astarix#3

Relevant links:

• https://github.com/eth-sri/astarix
• https://link.springer.com/chapter/10.1007/978-3-031-04749-7_22
• https://link.springer.com/chapter/10.1007/978-3-030-45257-5_7

SeGram

Reasons for failure:

• There is no implementation of CLI
• Using the tool requires manual changes to the source code to use desired input
• An initial seed for the alignment is required to be specified manual
• The output format is not specified, and it is currently impossible to run the tool
• It is thus impossible to fairly compare with other methods

Relevant links:

• https://github.com/CMU-SAFARI/SeGraM
• https://people.inf.ethz.ch/omutlu/pub/SeGraM_genomic-sequence-mapping-universal-accelerator_isca22.pdf