wvictor14 / planet Goto Github PK

Some users reporting that estimates can vary greatly when subsets of features are missing. e.g. Missing 6 of the 59 rs probes can significantly impact prediction accuracy

option 1 - change default behavior to throw stop instead of warn (current default behaviour). Perhaps provide suggestions in documentation on what can be done when it is impossible to collect all features (e.g. PCA).

Hi Victor,

I was trying to use predictEthnicity to predict the ethnicity from EPIC array data but resulted table showed all "NA". I did checked my data using the ethnicityCpGs function and it returned a TRUE. Could you suggest what might be the reason of not getting the result?

Thansk!

Requiring glmnet as a dependency for pl_infer_ethnicity is unnecessary. Simply multiplying the coefficients by the features vector should work fine.

Refer to caret #978 for code

Installing on R 3.6.0 fails:
r-lib/remotes#374 (comment)

Installing on R 3.5.3, 3.5.1 works fine (only ones I tested).

Current workaround is to run:

withr::with_envvar(c(R_REMOTES_NO_ERRORS_FROM_WARNINGS="true"), 
  remotes::install_github('wvictor14/planet')
)

README is long

• Make README short, and make a vignette that is roughly equivalent to the current README

automatically update website with R package updates

https://victor.rbind.io/planet/

add @iciarfernandez preeclampsia predictor

Address #16 #11

• pl_infer_ethnicity (in progress)

• pl_infer_age

• pl_cell_cpgs_third / pl_cell_cpgs_first

• add badge

Ideally some .idats are used to showcase example usage. Challenge is that .idat files are >5 mb each which will exceed github's limit. That means data needs to be hosted elsewhere.

• Consider submitting to bioconductor
• Consider reading data into R using GEOquery
• Other file hosting services?

          One more thing - it was discussed at lab meeting that instead of `Ambiguous`, it should say `Other` in the labels that `predictEthnicity` outputs, since the tool can only calculate 3 ancestries but there are other ancestries out there (+ ancestry is a continuum) so in reality samples being called ambiguous may just be mixed or from an ancestry other than African/Asian/European. Let me know what you think and if you agree I'm happy to change that myself too!

Originally posted by @iciarfernandez in #19 (comment)