Closed benmwebb closed 7 years ago
@Pellarin @cgreenberg Any idea where these trajectories can be found?
Per Riccardo, most likely they are (somewhere) in /salilab/park1/pellarin/mediator.tar.xz
Yes, if we can untar that 30GB file somewhere, I can definitely look into it!
See /salilab/park1/pellarin/no-backup/mediator/
.
in /salilab/park1/pellarin/no-backup/mediator/for_publication/analysis
the directories clustering clustering_half1 clustering_half2 clustering_jackknifing
contains the clustering analysis published in the paper. In each directory look the clustering.py
file which contains the path of the original trajectories. For instance, for clustering/clustering.py
you have:
"../../run20/modeling.split_density.scale_radii.two_bayesian_classes-1.1",
"../../run20/modeling.split_density.scale_radii.two_bayesian_classes-1.2",
"../../run20/modeling.split_density.scale_radii.two_bayesian_classes-1.3",
"../../run20/modeling.split_density.scale_radii.two_bayesian_classes-1.4",
"../../run28/modeling.split_density.scale_radii.two_bayesian_classes-1.5",
"../../run28/modeling.split_density.scale_radii.two_bayesian_classes-1.6",
"../../run28/modeling.split_density.scale_radii.two_bayesian_classes-1.7",
"../../run28/modeling.split_density.scale_radii.two_bayesian_classes-1.8",
"../../run31/modeling.split_density.scale_radii.two_bayesian_classes-1.9",
"../../run31/modeling.split_density.scale_radii.two_bayesian_classes-1.10",
"../../run31/modeling.split_density.scale_radii.two_bayesian_classes-1.11",
"../../run31/modeling.split_density.scale_radii.two_bayesian_classes-1.12",
"../../run31/modeling.split_density.scale_radii.two_bayesian_classes-1.13",
"../../run31/modeling.split_density.scale_radii.two_bayesian_classes-1.14",
"../../run31/modeling.split_density.scale_radii.two_bayesian_classes-1.15",
"../../run31/modeling.split_density.scale_radii.two_bayesian_classes-1.16",
"../../run31/modeling.split_density.scale_radii.two_bayesian_classes-1.17",
"../../run31/modeling.split_density.scale_radii.two_bayesian_classes-1.18",
"../../run31/modeling.split_density.scale_radii.two_bayesian_classes-1.19",
"../../run31/modeling.split_density.scale_radii.two_bayesian_classes-1.20"
Perfect, thanks Riccardo! I'll archive these directories at Zenodo - they're not too big.
Awesome! Thanks!
Normally I would expect to see one or more directories with output models (prior to clustering), for example in an
output
subdirectory (here is the equivalent for the exosome modeling) but I don't see that for Mediator.I'm sure that these trajectories are too big to fit in a GitHub repository, which is why they weren't uploaded in the first place. But I'd still like to have them for reference. We can upload them to Zenodo or (worst case) put them on the Sali lab's FTP site. Where are they?