Skip to content

method re-selection - improves runtime and memory usage #1

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
rl3328 merged 1 commit into from
Jan 16, 2026
+20 −19
Merged

method re-selection - improves runtime and memory usage #1

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
39 changes: 20 additions & 19 deletions code/pecotmr_integration/twas_ctwas.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -577,31 +577,32 @@
"\n",
"\n",
" # function to update method selection\n",
" update_twas_method <- function(twas_df){\n",
" update_twas_method <- function(twas_df, rsq_cutoff =${rsq_cutoff}, rsq_pval_cutoff = ${rsq_pval_cutoff}){\n",
" if (nrow(twas_df) == 0) { \n",
" return(twas_df)\n",
" } \n",
" twas_df$method_selected_original <- twas_df$is_selected_method\n",
" twas_df_colnames <- colnames(twas_df)\n",
" twas_df$gene_context <- paste0(twas_df$molecular_id, \"_\", twas_df$context)\n",
" rs_update <- do.call(rbind, lapply(unique(twas_df$gene_context), function(gene_context){\n",
" do.call(rbind, lapply(unique(twas_df$gwas_study), function(study){\n",
" df <- twas_df[twas_df$gene_context==gene_context & twas_df$gwas_study==study,,drop=FALSE]\n",
" df_colnames <- colnames(df)\n",
" all_imputable <- df[df$rsq_cv>=${rsq_cutoff} & df$pval_cv<=${rsq_pval_cutoff} & !is.na(df$twas_z),,drop=FALSE]\n",
" if (nrow(all_imputable)<=1 || all(is.na(all_imputable$twas_z))) {\n",
" return(df) # gene-context pair does not having any other imputable model available.\n",
" } else {\n",
" message (\"Updating method selection for \", gene_context, \"-\", study, \".\")\n",
" selected_method_update <- all_imputable[order(-all_imputable$rsq_cv),]$method[1]\n",
" df$selected_method_update <- df$method==selected_method_update\n",
" df$is_selected_method <- df$selected_method_update\n",
" df <- df[, df_colnames]\n",
" return(df)\n",
" }\n",
" }))\n",
" }))\n",
" return(rs_update[, twas_df_colnames])\n",
" fix_sel <- twas_df$is_selected_method & (is.na(twas_df$twas_z) | is.na(twas_df$twas_pval) | is.infinite(twas_df$twas_pval) |\n",
" is.null(twas_df$twas_z) | is.null(twas_df$twas_pval))\n",
" if (!any(fix_sel)) return(twas_df) # If nothing to fix, return fast\n",
" all_group <- paste(twas_df$gene_context, twas_df$gwas_study, sep = \" - \") #gene-context-study group\n",
" fix_groups <- unique(all_group[fix_sel])\n",
"\n",
" for (fixgroup in fix_groups) {\n",
" df_idxs <- which(all_group == fixgroup) # idx of gene-context-study group that needs replacement of method selected\n",
" if (length(df_idxs) == 1) next\n",
" candidate_idxs <- df_idxs[!is.na(twas_df$twas_z[df_idxs]) & !is.na(twas_df$twas_pval[df_idxs]) & !is.infinite(twas_df$twas_z[df_idxs]) & \n",
" !is.null(twas_df$twas_z[df_idxs]) & !is.null(twas_df$twas_pval[df_idxs]) &\n",
" twas_df$rsq_cv[df_idxs] >= rsq_cutoff & twas_df$pval_cv[df_idxs] <= rsq_pval_cutoff] \n",
" if (length(candidate_idxs) <= 0 || all(is.na(candidate_idxs))) next\n",
" best_i <- candidate_idxs[which.max(twas_df$rsq_cv[candidate_idxs])]\n",
" best_method <- twas_df$method[best_i]\n",
" twas_df$is_selected_method[df_idxs] <- (twas_df$method[df_idxs] == best_method)\n",
" message (\"Updating method selection for \", fixgroup, \".\")\n",
" }\n",
" return(twas_df[, twas_df_colnames])\n",
" }\n",
"\n",
" # Load metadata and configuration - let these fail if there are issues\n",
Expand Down
Loading