predict_for_linearization

Author

Saideep Gona

Published

November 16, 2023

Code

library(data.table)
library(glmnet)
library(glue)
library(tidyverse)

Code

X_path <- "/beagle3/haky/users/saideep/projects/aracena_modeling/linearization/HG00096.txt"

X <- as.matrix(data.table::fread(X_path))

glmnet_model_path <- "/beagle3/haky/users/saideep/projects/aracena_modeling/elastic_net/trained_eln_RNAseq_NI_from_HDF5_mean_log.linear.rds"
glmnet_model <- readRDS(glmnet_model_path)

glmnet_predictions <- predict(glmnet_model, X, s = "lambda.min", type="response")

write

function (x, file = "data", ncolumns = if (is.character(x)) 1 else 5, 
    append = FALSE, sep = " ") 
cat(x, file = file, sep = c(rep.int(sep, ncolumns - 1), "\n"), 
    append = append)
<bytecode: 0x5641f3d9f930>
<environment: namespace:base>

Code

# inds <- read_file("/beagle3/haky/users/saideep/github_repos/Daily-Blog-Sai/posts/2023-11-16-linearization/individuals.txt")
# inds <- gsub("\n", "", inds)
# inds <- strsplit(inds,split="\t")


# linear_dir <- "/beagle3/haky/users/saideep/projects/aracena_modeling/linearization"

# for (ind in inds[[1]]) {
#     X_path <- glue("/beagle3/haky/users/saideep/projects/aracena_modeling/linearization/{ind}.txt")
#     print(ind)
#     if (file.exists(X_path)) {
#         X <- as.matrix(data.table::fread(X_path))

#         glmnet_model_path <- "/beagle3/haky/users/saideep/projects/aracena_modeling/elastic_net/trained_eln_RNAseq_NI_from_HDF5_mean_log.linear.rds"
#         glmnet_model <- readRDS(glmnet_model_path)

#         glmnet_predictions <- as.data.frame(predict(glmnet_model, X, s = "lambda.min", type="response"))

#         write_delim(glmnet_predictions, glue("/beagle3/haky/users/saideep/projects/aracena_modeling/linearization/{ind}_preds.txt"), delim = "\t")

#     } else {
#         print(glue({"{ind} missing"}))
#     }
# }