These are some pangenomes created using Minigraph-Cactus, along with all steps used to produce them.
The commits used can be found in the logs in the Output links. Data is provided in HAL, vg, GFA and VCF formats. Please read about VCF output before using it.
Please cite the Minigraph-Cactus paper when using these pangenomes. For the HPRC pangenomes, please also cite the HPRC Paper.
- These pangenomes are provided "as-is" and, unless published on, are here merely to serve as examples as we've used them for debugging and testing but haven't necessarily spent much time validating them.
- The Minigraph-Cactus Pangenome pipeline's interface has seen many changes (especially prior to publication of the methdod). It's possible that the given commands do not work with the latest version of Cactus! Try to look at the most recent example(s) if you are having issues with the interface.
- The
30-mouse-pg-2022-09-23
pangenome is nearly 40% Ns due, apparently, to gappy input assemblies. This should be taken into account when using it. How best to handle Ns is still an area of future investigation. - The commands given below (upt to early 2023, when we switched to SLURM) to construct these graphs were run from Toil AWS leader instances. If you are not running on an AWS/EC2 cluster and want to reproduce the graphs, then you will need to adapt the commands to your environment. To run locally remove the AWS-specific options like
--batchSystem
,--defaultPreemptable
,--nodeType
,--nodeStorage
,--maxNodes
,--betaInertia
,--targetTime
and--provisioner
and use a local (non-s3) jobstore (like./jobstore
).
Creating an HPRC seqfile and running the current pipeline on it is described here.