#ژنتیک
#نکات_استراخان
#نکته۷۵
کانال آموزشی آرش صفرزاده (@GeneticsAS
Please open Telegram to view this post
VIEW IN TELEGRAM
❤5❤🔥4🙏2⚡1
Linear regression
Bias and Variance
Cross Validation
Ridge regression
LASSO
df <- dataset1[rownames(dataset1) %in% hubs, ]
df <- t(df)
df <- data.frame(df)
df$Status <- c(rep("Tumor", 25), rep("Normal", 25))
x <- as.matrix(df[, -ncol(df)])
View(x)
y <- c(rep(1,25), rep(0, 25))
تو کد بالا، hubs یک وکتور حاوی اسامی هابژنها و dataset1 هم دیتای بیانی نرمالایزشده هست.
library(glmnet)
cv_fit <- cv.glmnet(x = x,
y = y,
alpha = 1,
family = "binomial",
nfolds = 10,
type.measure = "deviance")
plot(cv_fit)
best_lambda <- cv_fit$lambda.min
lasso_model <- glmnet(x = as.matrix(df[, -30]),
y = labels,
alpha = 1,
lambda = best_lambda,
family = "binomial")
coef(lasso_model)
selected_genes <- rownames(coef(lasso_model))[coef(lasso_model)[,1] != 0]
selected_genes <- selected_genes[-1]
# Fit the full LASSO model (not cross-validation) to extract paths
fit <- glmnet(x = as.matrix(df[, -30]),
y = labels,
alpha = 1,
family = "binomial")
plot(fit,
xvar = "lambda",
label = TRUE,
main = "LASSO Coefficient Paths",
cex.main = 1)
Machine learning and WGCNA reveal the PVT1/miR-143–3p/CDK1 ceRNA axis as a key regulator in NSCLC
#بیوانفورماتیک
کانال آموزشی آرش صفرزاده (@GeneticsAS
Please open Telegram to view this post
VIEW IN TELEGRAM
❤11❤🔥4😍2👍1👌1🍓1
Anonymous Quiz
48%
TGFβ
15%
Wnt
11%
SHH
25%
Notch-delta
😎9❤6❤🔥2👍2🙏1💯1😘1
#تست_روز
کانال آموزشی آرش صفرزاده (@GeneticsAS
Please open Telegram to view this post
VIEW IN TELEGRAM
❤9❤🔥3🙏2🏆2💯1
Decision and Classification Trees (Part 1)
Decision and Classification Trees (Part 2)
Random Forest (Part 1)
Random Forest (Part 2)
Random Forest in R
df <- dataset1[rownames(dataset1) %in% hubs, ]
df <- t(df)
df <- data.frame(df)
df$Status <- c(rep("Tumor", 25), rep("Normal", 25))
x <- as.matrix(df[, -ncol(df)])
View(x)
y <- c(rep(1,25), rep(0, 25))
تو کد بالا، hubs یک وکتور حاوی اسامی هابژنها و dataset1 هم دیتای بیانی نرمالایزشده هست.
library(randomForest)
rf_model <- randomForest(Status ~ .,
data = df,
importance = TRUE,
ntree = 120)
print(rf_model)
plot(rf_model)
importance_vals <- importance(rf_model)
varImpPlot(rf_model)
important_genes_rf <- importance_vals[order(importance_vals[, "MeanDecreaseGini"], decreasing = TRUE), ]
روش اول: آستانه روی MeanDecreaseAccuracy (Z-score):
z_scores <- scale(important_genes_rf[, "MeanDecreaseAccuracy"])
selected_genes <- names(z_scores[z_scores > 1])
selected_genes
روش دوم: انتخاب ژنهای برتر (Top genes)؛ مثلا ده ژن برتر:
top_genes_rf <- rownames(important_genes_rf)[1:10]
Machine learning and WGCNA reveal the PVT1/miR-143–3p/CDK1 ceRNA axis as a key regulator in NSCLC
#بیوانفورماتیک
کانال آموزشی آرش صفرزاده (@GeneticsAS
Please open Telegram to view this post
VIEW IN TELEGRAM
❤9❤🔥2👌2💯1🏆1
Anonymous Quiz
22%
Conceal
56%
Unearth
10%
Fabricate
13%
Disintegrate
❤7😎5👍3👏1
#تست_زبان
کانال آموزشی آرش صفرزاده (@GeneticsAS
Please open Telegram to view this post
VIEW IN TELEGRAM
❤9🙏3❤🔥2🏆2👍1
@GeneticsAS Cancer Genetics Table Part 1.pdf
295.1 KB
GeneticsASکانال آموزشی آرش صفرزاده (@GeneticsAS
Please open Telegram to view this post
VIEW IN TELEGRAM
1❤17😍5❤🔥3🙏3👍1🏆1
Decision and Classification Trees (Part 1)
Decision and Classification Trees (Part 2)
Random Forest (Part 1)
Random Forest (Part 2)
df <- dataset1[rownames(dataset1) %in% hubs, ]
df <- t(df)
df <- data.frame(df)
df$Status <- c(rep("Tumor", 25), rep("Normal", 25))
x <- as.matrix(df[, -ncol(df)])
View(x)
y <- c(rep(1,25), rep(0, 25))
تو کد بالا، hubs یک وکتور حاوی اسامی هابژنها و dataset1 هم دیتای بیانی نرمالایزشده هست.
library(Boruta)
df$Status <- as.factor(df$Status)
set.seed(123)
boruta_result <- Boruta(Status ~ .,
data = df,
doTrace = 2,
maxRuns = 500)
par(mar = c(5, 5, 5, 5))
plot(boruta_result,
las = 2,
cex.axis = 0.6,
main = "Boruta Feature Importance")
final_boruta <- TentativeRoughFix(boruta_result)
selected_boruta <- getSelectedAttributes(final_boruta,
withTentative = FALSE)
print(selected_boruta)
Machine learning and WGCNA reveal the PVT1/miR-143–3p/CDK1 ceRNA axis as a key regulator in NSCLC
#بیوانفورماتیک
کانال آموزشی آرش صفرزاده (@GeneticsAS
Please open Telegram to view this post
VIEW IN TELEGRAM
❤8❤🔥2👌2😍2👏1🍓1💋1
@GeneticsAS Cancer Genetics Table Part 2.pdf
290.5 KB
پارت اول
GeneticsASکانال آموزشی آرش صفرزاده (@GeneticsAS
Please open Telegram to view this post
VIEW IN TELEGRAM
❤7🙏4❤🔥2👍1🏆1🍓1
Anonymous Quiz
10%
اثر خنثی
44%
اثر افزایشی
25%
اثر کاهشی
20%
اثر کاهشی در شرایط خاص
😱6❤4👍3❤🔥1👏1🍓1
#تست_روز
کانال آموزشی آرش صفرزاده (@GeneticsAS
Please open Telegram to view this post
VIEW IN TELEGRAM
❤6❤🔥3🙏2👏1🏆1🍓1
@GeneticsAS - English Passages.pdf
268 KB
GeneticsAS#زبان
#تست_زبان
کانال آموزشی آرش صفرزاده (@GeneticsAS
Please open Telegram to view this post
VIEW IN TELEGRAM
❤7❤🔥3👏1🙏1🍓1😎1
#ژنتیک
#نکات_جرد
کانال آموزشی آرش صفرزاده (@GeneticsAS
Please open Telegram to view this post
VIEW IN TELEGRAM
❤6❤🔥2👏2👍1🏆1🍓1
Anonymous Quiz
66%
FTO
19%
MC4R
6%
IRX
9%
ALDH
❤8❤🔥3✍2😢2🏆1🍓1
#تست_روز
کانال آموزشی آرش صفرزاده (@GeneticsAS
Please open Telegram to view this post
VIEW IN TELEGRAM
❤10👏2🙏2💯1🏆1🍓1
#اخبار
#کارشناسی_ارشد
کانال آموزشی آرش صفرزاده (@GeneticsAS
Please open Telegram to view this post
VIEW IN TELEGRAM
❤11👍3❤🔥2👌1
Decision and Classification Trees (Part 1)
Decision and Classification Trees (Part 2)
Random Forest (Part 1)
Random Forest (Part 2)
Gradient Boost (Part 1)
Gradient Boost (Part 2)
Gradient Boost (Part 3)
Gradient Boost (Part 4)
XGBoost (Part 1)
XGBoost (Part 2)
XGBoost (Part 3)
XGBoost (Part 4)
df <- dataset1[rownames(dataset1) %in% hubs, ]
df <- t(df)
df <- data.frame(df)
x <- as.matrix(df[, -ncol(df)])
y <- c(rep(1,25), rep(0, 25))
library(xgboost)
library(caret)
library(Matrix)
dtrain <- xgb.DMatrix(data = x,
label = y)
class(dtrain)
params <- list(booster = "gbtree",
objective = "binary:logistic",
eval_metric = "logloss")
xgb_model <- xgb.train(params = params,
data = dtrain,
nrounds = 100,
verbose = 0)
importance <- xgb.importance(model = xgb_model)
print(importance)
xgb.plot.importance(importance_matrix = importance,
top_n = 29,
col = col)
top_markers <- importance$Feature[1:5]
print(top_markers)
Machine learning and WGCNA reveal the PVT1/miR-143–3p/CDK1 ceRNA axis as a key regulator in NSCLC
#بیوانفورماتیک
کانال آموزشی آرش صفرزاده (@GeneticsAS
Please open Telegram to view this post
VIEW IN TELEGRAM
❤7👏3😍2💯2🏆1
#اخبار
#کارشناسی_ارشد
کانال آموزشی آرش صفرزاده (@GeneticsAS
Please open Telegram to view this post
VIEW IN TELEGRAM
👍7❤3🙏2👌1
Anonymous Quiz
38%
TWIST
26%
YAP1
23%
ZNF3
13%
POMC
❤8😎4❤🔥2🔥2🙏1
#تست_روز
کانال آموزشی آرش صفرزاده (@GeneticsAS
Please open Telegram to view this post
VIEW IN TELEGRAM
❤8🙏3🏆2❤🔥1🍓1