b9c41cec68f33aec1a64d8cc709af9afc15155fa
mspeir
Fri Sep 25 22:49:03 2026 -0700
hg38 cons241way: point the phyloP tree model text at the directory, refs #34803
The link went to phyloP241way/hg38.phyloP241way.mod, which does not exist and
never has under that name. The models live in cactus241way/phyloP241way, and
there are three of them rather than one: a general model and separate chrX and
chrY models. Link the directory and say so.
Co-Authored-By: Claude Opus 5 (1M context)
This same tree model was used in the phyloP calculations; however, the -background frequencies were modified to maintain reversibility. -The resulting tree model: -all species. +background frequencies were modified to maintain reversibility. The resulting +models are in the +phyloP241way download directory as .mod files: a general +model, plus separate models for chrX and chrY.
The phyloP program supports several different methods for computing p-values of conservation or acceleration, for individual nucleotides or larger elements ( http://compgen.cshl.edu/phast/). Here it was used to produce separate scores at each base (--wig-scores option), considering all branches of the phylogeny rather than a particular subtree or lineage (i.e., the --subtree option was not used). The scores were computed by performing a likelihood ratio test at each alignment column (--method LRT), and scores for both conservation and acceleration were produced (--mode CONACC).