search

smirarab / ASTRAL

Accurate Species TRee ALgorithm
Apache License 2.0
234 stars 68 forks source link

Help in interpreting output log and species tree #52

Closed matiaskohler closed 4 years ago

matiaskohler commented 5 years ago

I ran Astral 5.6.2. My input file was a set of 72 gene-trees from RAxML "bestTrees". I ran just with the standard options -i -o and 2> to save the log.

I got the normalized score and the final normalized quartet score as the same value: 0.686484593837535 How do I interpret this value in a biological point of view?

When I open the output tree of Astral in FigTree, it ask me to set a name for the values that are labelled on node/braches of the tree. Are these values the CU that Astral estimate?

I have the value 1.0 for the majority of nodes, and 0.4-0.6 for a few nodes.
What the lower values could be indicating me in a biological point of view?

smirarab commented 5 years ago

Please see: https://arxiv.org/abs/1904.03826 Values you are referring to are localPP not CU branch lengths. Quartet score is the percentage of quartet trees missing from gene trees. It is a measure of gene tree discordance. 1.0 means no gene tree discordance. 0.3333 is extreme gene tree discordance (random trees).

On Thu, Jun 20, 2019 at 1:24 PM Matias Köhler notifications@github.com wrote:

I ran Astral 5.6.2. My input file was a set of 72 gene-trees from RAxML "bestTrees". I ran just with the standard options -i -o and 2> to save the log.

I got the normalized score and the final normalized quartet score as the same value: 0.686484593837535 How do I interpret this value in a biological point of view?

When I open the output tree of Astral in FigTree, it ask me to set a name for the values that are labelled on node/braches of the tree. Are these values the CU that Astral estimate?

I have the value 1.0 for the majority of nodes, and 0.4-0.6 for a few nodes. What the lower values could be indicating me in a biological point of view?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/smirarab/ASTRAL/issues/52?email_source=notifications&email_token=AAGJXOAA3TORJWO5A5P52ULP3PRPTA5CNFSM4HZ5HXD2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4G2ZTPUA, or mute the thread https://github.com/notifications/unsubscribe-auth/AAGJXOEA3ELKT7UQAV7RBADP3PRPTANCNFSM4HZ5HXDQ .

-- Siavash Mirarab