lamho86 / phylolm

GNU General Public License v2.0
30 stars 12 forks source link

How to explain the phylolm results #19

Closed lychen83 closed 4 years ago

lychen83 commented 5 years ago

Hello,

I am trying to find out the genes with expression level correlated to the trait (coded as 0 and 1) with phylolm. I have one question. When I run the phylolm using following command: library('ape') library('phylolm') setwd("path) getwd() samples = read.table("expression_trait_file",row.names=1) tree = read.nexus("tree", tree.names = NULL, force.multi = FALSE) fit = phyloglm(formula=pigment~tpm, data=samples, phy=tree, method = ("logistic_MPLE"), boot = 0) summary(fit) coef(fit) vcov(fit)

I got results as following:

summary(fit)

Call: phyloglm(formula = pigment ~ tpm, data = samples, phy = tree, method = ("logistic_MPLE"), boot = 0) AIC logLik Pen.logLik 44.35 -19.17 -16.78

Method: logistic_MPLE Mean tip height: 0.3558203 Parameter estimate(s): alpha: 2.787424

Coefficients: Estimate StdErr z.value p.value
(Intercept) -2.091304 1.114385 -1.8766 0.06057 . tpm 0.142301 0.099049 1.4367 0.15081

Signif. codes: 0 ‘’ 0.001 ‘’ 0.01 ‘’ 0.05 ‘.’ 0.1 ‘ ’ 1

Note: Wald-type p-values for coefficients, conditional on alpha=2.787424

coef(fit) (Intercept) tpm -2.0913038 0.1423013 vcov(fit) (Intercept) tpm (Intercept) 1.24185403 -0.062150513 tpm -0.06215051 0.009810609

Could you kindly me let me know how I can know if the gene expression is significantly correlated to the trait (coded a 0 and 1) from the output? In other words, could you kindly tell me how to explain the results? Thank you very much.

Best regards,

Lingyun

cecileane commented 5 years ago

your p-value is 0.15081: it seems that there is no evidence of a correlation.