Table 2 Blob detection using TINNiK for simulated data for \({\mathcal {N}}^+\) of Fig. 1 (L). Entries give ranges for \(\alpha \) on which the full tree of blobs, and individual blobs, are correctly inferred with the T3 test. Interval endpoints are approximate, with dashes indicating the correct multifurcation is never inferred. For all analyses, \(\beta =1\)
From: TINNiK: inference of the tree of blobs of a species network under the coalescent model
Number n of gene trees | Range of \(\alpha \) values, for blob detection. (\(\beta = 1\).) | |||
|---|---|---|---|---|
Tree of Blobs | ||||
Fully correct | B-blob detected | C-blob detected | D-blob detected | |
Low ILS: \(k=2\). | ||||
10000 | \([ 10^{-170}, \, 0.01 ]\) | \([ 10^{-170}, \, 0.01]\) | \([ 10^{-300}, \, 0.01 ]\) | \([ 10^{-170}, \, 0.01 ]\) |
1000 | \([ 10^{-15},\, 10^{-4}]\) | \([ 10^{-16},\, 10^{-4} ]\) | \([ 10^{-49},\, 0.01]\) | \([ 10^{-15}, 10^{-4},]\) |
500 | \([ 10^{-6},\, 0.001]\) | \([ 10^{-7},\, 0.001]\) | \([ 10^{-25},\, 0.01]\) | \([ 10^{-6},\, 0.001]\) |
300 | – | \([ 10^{-6},\, 10^{-5}]\) | \([ 10^{-17},\, 0.01]\) | – |
300 | – | – | – | – |
Moderate ILS: \(k=1\). | ||||
10000 | \([ 10^{-55} ,\ 0.001] \) | \([ 10^{-55} ,\ 0.001]\) | \([ 10^{-216}, \, 0.001]\) | \([ 10^{-55} ,\ 0.001]\) |
1000 | \([ 10^{-7},\, 0.001]\) | \([ 10^{-7},\, 0.001]\) | \([ 10^{-23} ,\, 0.01]\) | \([ 10^{-7},\, 0.001]\) |
500 | \([ 10^{-4} ,\, 0.005]\) | \([ 10^{-4} ,\, 0.008]\) | \([1\times 10^{-13}, \, 0.01 ]\) | \([ 10^{-4} ,\, 0.005]\) |
300 | – | – | \([ 10^{-7}, \, 0.003]\) | – |
300 | – | – | – | – |
High ILS: \(k=0.5\). | ||||
10000 | \([ 10^{-12},\, 0.001]\) | \([ 10^{-12},\, 0.001]\) | \([ 10^{-88},\, 0.001]\) | \([ 10^{-17},\, 0.001]\) |
1000 | – | – | \([ 10^{-8},\, 0.001]\) | – |
500 | – | – | \([ 10^{-4},\, 0.001]\) | – |
300 | – | – | – | – |