dataset.md

Dataset

A minimal example dataset is provided called toy1, which can be downloaded with:

rebar dataset download --name toy1 --tag custom --output-dir dataset/toy1

Mandatory

A rebar dataset consists of two mandatory parts:

1. reference.fasta: The reference genome.

   >Reference
   AAAAAAAAAAAAAAAAAAAA
   

2. populations.fasta: Known populations (ex. clades, lineages) aligned to the reference.

   >A
   CCCCCCAACCCCCCCCCCCC
   >B
   TTTTTTTTTTTTTTTTTTAA
   >C
   AAGGGGGGGGGGGGGGGGGG
   >D
   CCCCCCAACCCTTTTTTTAA
   >E
   AAGCCCAACCCTTTTTTTAA
   

Optional

The following are optional components:

1. annotations.tsv: A table of genome annotations to add to the plot.

   gene	abbreviation	start	end
   Gene1	g1	1	3
   Gene2	g2	12	20

2. phylogeny.json: A phylogenetic graph which provides prior information about the evolutionary history. This is particularly useful if populations in populations.fasta are internal nodes or known recombinants.

   For example, an evolutionary history, in which population D is a recombinant, and E is a recursive recombinant (it has a parent that is also a recombinant).

   graph LR;
       root-->A;
       root-->B;
       root-->C;
       A-->D;
       B-->D;
       C-->E;
       D-->E;
   
   style D fill:#00758f
   style E fill:#00758f
   

   Loading

   Is represented by the following phylogenetic graph:

   {
     "graph": {
       "nodes": ["root", "A", "B", "C", "D", "E" ],
       "edge_property": "directed",
       "edges": [
         [ 0, 1, 1 ],
         [ 0, 2, 1 ],
         [ 0, 3, 1 ],
         [ 1, 4, 1 ],
         [ 2, 4, 1 ],
         [ 4, 5, 1 ],
         [ 3, 5, 1 ]
       ]
     }
   }

   Where nodes are the list of node names in the tree (internal and external), and edges are the branches between nodes. For example, the edge [0, 1, 1] connects node index 0 ("root") to node index 1 ("A") with a branch length of 1. Please note that branch lengths are not currently used in rebar's algorithm.

3. edge_cases.json: A list of rebar arguments to apply only to a particular population.

   In the dataset, E is a recursive recombinant between population C and recombinant D. However, we could instead force it to be a recombinant between A, B, and C with the following parameters:

   [
     {
       "population": "E",
       "parents": [ "A", "B", "C"],
       "knockout": null,
       "mask": [0, 0],
       "max_iter": 3,
       "max_parents": 3,
       "min_parents": 3,
       "min_consecutive": 3,
       "min_length": 3,
       "min_subs": 1,
       "naive": false
     }
   ]

   Default	Edge Case