tanu/LSHTM_analysis
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

ML scripts: {'n_jobs': os.cpu_count() }

Browse source
This commit is contained in:
Tanushree Tunstall 2022-07-02 10:20:40 +01:00
parent 11af00f1db
commit b8653c6afe
4 changed files with 16 additions and 12 deletions
Download patch file Download diff file Expand all files Collapse all files

3
scripts/ml/ml_functions/FS.py
View file

@ -76,8 +76,7 @@ import argparse
import re import re
##################################### #####################################
rs = {'random_state': 42} rs = {'random_state': 42}
njobs = {'n_jobs': 10} njobs = {'n_jobs': os.cpu_count() } # the number of jobs should equal the number of CPU cores
scoring_fn = ({ 'mcc' : make_scorer(matthews_corrcoef) scoring_fn = ({ 'mcc' : make_scorer(matthews_corrcoef)
, 'fscore' : make_scorer(f1_score) , 'fscore' : make_scorer(f1_score)

2
scripts/ml/ml_functions/MultClfs.py
View file

@ -76,7 +76,7 @@ import argparse
import re import re
#%% GLOBALS #%% GLOBALS
rs = {'random_state': 42} rs = {'random_state': 42}
njobs = {'n_jobs': 10} njobs = {'n_jobs': os.cpu_count() } # the number of jobs should equal the number of CPU cores
scoring_fn = ({ 'mcc' : make_scorer(matthews_corrcoef) scoring_fn = ({ 'mcc' : make_scorer(matthews_corrcoef)
, 'fscore' : make_scorer(f1_score) , 'fscore' : make_scorer(f1_score)

2
scripts/ml/ml_functions/SplitTTS.py
View file

@ -41,7 +41,7 @@ import re
homedir = os.path.expanduser("~") homedir = os.path.expanduser("~")
#%% GLOBALS #%% GLOBALS
rs = {'random_state': 42} rs = {'random_state': 42}
njobs = {'n_jobs': 10} njobs = {'n_jobs': os.cpu_count() } # the number of jobs should equal the number of CPU cores
#%% Define split_tts function ################################################# #%% Define split_tts function #################################################
def split_tts(ml_input_data def split_tts(ml_input_data

21
scripts/plotting/lineage_plots_multipage.R
View file

@ -5,8 +5,8 @@ library(ggpubr)
library(svglite) library(svglite)
# for testing only # for testing only
gene="pncA" #gene="pncA"
drug="pyrazinamide" #drug="pyrazinamide"
lineage_plot=function(gene,drug){ lineage_plot=function(gene,drug){
lineage_filename=paste0(tolower(gene),"_merged_df2.csv") lineage_filename=paste0(tolower(gene),"_merged_df2.csv")
@ -84,7 +84,7 @@ lineage_plot=function(gene,drug){
#print (i) #print (i)
s_mut = plot_df[plot_df$mutationinformation == i,] s_mut = plot_df[plot_df$mutationinformation == i,]
s_tab = table(s_mut$lineage, s_mut$sensitivity) s_tab = table(s_mut$lineage, s_mut$sensitivity)
ft_pvalue_i = fisher.test(s_tab)$p.value ft_pvalue_i = fisher.test(s_tab, workspace=2000000)$p.value
plot_df$pval[plot_df$mutationinformation == i] <- ft_pvalue_i plot_df$pval[plot_df$mutationinformation == i] <- ft_pvalue_i
} }
plot_df$pvalR = round(plot_df$pval, 3) plot_df$pvalR = round(plot_df$pval, 3)
@ -131,12 +131,14 @@ lineage_plot=function(gene,drug){
# Do plots # Do plots
plot_pages = round(length(lin_muts)/25) plot_pages = round(length(lin_muts)/25)
if (plot_pages<1){plot_pages=1}
p_title = gene p_title = gene
res = 144 # SVG dots-per-inch res = 144 # SVG dots-per-inch
print(paste0('About to plot ', plot_pages, ' page(s).'))
sapply(1:plot_pages, function(page){ sapply(1:plot_pages, function(page){
print(paste0("Plotting page:", page)) print(paste0("Plotting page:", page))
svglite(paste0("/tmp/",drug,"-",page,".svg"), width=2048/res, height=1534/res) # old-school square 4:3 CRT shape 1.3:1 svglite(paste0("/tmp/",drug,"-",page,".svg"), width=2048/res, height=1534/res) # old-school square 4:3 CRT shape 1.33:1
print( print(
ggplot(plot_df2, aes(x = lineage ggplot(plot_df2, aes(x = lineage
, fill = sensitivity)) + , fill = sensitivity)) +
@ -169,17 +171,19 @@ lineage_plot=function(gene,drug){
# hardcoded list of drugs # hardcoded list of drugs
drugs = c(#"ethambutol", drugs = c(#"ethambutol",
#"isoniazid", "isoniazid",
"pyrazinamide", "pyrazinamide",
"rifampicin", "rifampicin",
"streptomycin", "streptomycin",
"cycloserine") #"cycloserine"
)
genes = c(#"embB", genes = c(#"embB",
#"katG", "katG",
"pncA", "pncA",
"rpoB", "rpoB",
"gid", "gid",
"alr") #"alr"
)
combo = data.frame(drugs, genes) combo = data.frame(drugs, genes)
#sapply(combo$drugs, function(x){print(c(x,combo[drugs==x,"genes"]))}) #sapply(combo$drugs, function(x){print(c(x,combo[drugs==x,"genes"]))})
@ -188,4 +192,5 @@ combo = data.frame(drugs, genes)
sapply(combo$drugs, function(drug){ sapply(combo$drugs, function(drug){
gene=combo[drugs==drug,"genes"] gene=combo[drugs==drug,"genes"]
lineage_plot(gene,drug) lineage_plot(gene,drug)
print(c(gene,drug))
}) })