bio-rna-quantification-count-matrix-qc

Quality control and exploration of RNA-seq count matrices before differential expression. Check for outliers, batch effects, and sample relationships. Use when assessing count matrix quality before DE analysis.

الانتقال إلى التثبيت

المستودع: mdbabumiamssm/LLMs-Universal-Life-Science-and-Clinical-Skills-
آخر نشاط في المصدر: ٤ فبراير ٢٠٢٦ في ١٩:٥٥
لغة SKILL.md المكتشفة: الإنجليزية
النجوم: ٣١
التفرعات: ٨

خيارات التثبيت

يُحدَّد Prompt الذي يراجع المصدر أولًا بشكل افتراضي. يمكنك التبديل إلى أمر مباشر أو تنزيل نسخة محلية.

التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.

يتجاوز الأمر المباشر Prompt المخصّص للمراجعة. افحص المصدر قبل تشغيله.

npx skills add https://github.com/mdbabumiamssm/LLMs-Universal-Life-Science-and-Clinical-Skills- --skill bio-rna-quantification-count-matrix-qc

يبقى الأمر في سطر واحد. مرّر أفقيًا لمراجعته كاملًا قبل النسخ.

تفضّل نسخة محلية؟ نزّل الملفات المتاحة حاليًا لدى SkillsMP.

مراجعة ملفات المصدر

اقرأ SKILL.md وأي ملفات مرافقة يعرضها SkillsMP قبل أن تقرر التثبيت.

مستكشف الملفات

4 ملفات

SKILL.md

تعليمات المصدر · معاينة للقراءة فقط

العودة إلى البحث العودة إلى التثبيت

import pandas as pd
import numpy as np

counts = pd.read_csv('count_matrix.csv', index_col=0)
metadata = pd.read_csv('sample_info.csv', index_col=0)

# Total counts per sample
colSums(counts(dds))

# Genes detected per sample
colSums(counts(dds) > 0)

# Counts summary
summary(colSums(counts(dds)))

total_counts = counts.sum()
genes_detected = (counts > 0).sum()

print('Total counts per sample:')
print(total_counts)
print('\nGenes detected:')
print(genes_detected)

# Remove genes with low counts across samples
keep <- rowSums(counts(dds)) >= 10
dds <- dds[keep, ]

# More stringent: at least N samples with count >= M
keep <- rowSums(counts(dds) >= 10) >= 3
dds <- dds[keep, ]

min_counts = 10
min_samples = 3

gene_filter = (counts >= min_counts).sum(axis=1) >= min_samples
counts_filtered = counts[gene_filter]

# Variance stabilizing transformation
vsd <- vst(dds, blind = TRUE)

# Or regularized log (slower, better for small n)
rld <- rlog(dds, blind = TRUE)

# Get transformed values
vst_matrix <- assay(vsd)

from sklearn.preprocessing import StandardScaler

cpm = counts * 1e6 / counts.sum()
log_cpm = np.log2(cpm + 1)

library(pheatmap)

# Sample correlation heatmap
sample_cor <- cor(assay(vsd))
pheatmap(sample_cor, annotation_col = coldata)

# Sample distance heatmap
sample_dist <- dist(t(assay(vsd)))
pheatmap(as.matrix(sample_dist), annotation_col = coldata)

import seaborn as sns
import matplotlib.pyplot as plt

sample_cor = log_cpm.corr()
sns.clustermap(sample_cor, annot=True, cmap='RdBu_r', center=0.9,
               vmin=0.8, vmax=1.0)
plt.savefig('sample_correlation.png')

# PCA plot
plotPCA(vsd, intgroup = 'condition')

# Custom PCA
pca <- prcomp(t(assay(vsd)))
pca_df <- data.frame(PC1 = pca$x[,1], PC2 = pca$x[,2],
                     condition = coldata$condition)

library(ggplot2)
ggplot(pca_df, aes(PC1, PC2, color = condition)) +
    geom_point(size = 3) +
    geom_text(aes(label = rownames(pca_df)), vjust = -0.5)

from sklearn.decomposition import PCA

pca = PCA(n_components=2)
pca_result = pca.fit_transform(log_cpm.T)

plt.figure(figsize=(8, 6))
for condition in metadata['condition'].unique():
    mask = metadata['condition'] == condition
    plt.scatter(pca_result[mask, 0], pca_result[mask, 1], label=condition)
plt.xlabel(f'PC1 ({pca.explained_variance_ratio_[0]:.1%})')
plt.ylabel(f'PC2 ({pca.explained_variance_ratio_[1]:.1%})')
plt.legend()
plt.savefig('pca_plot.png')

# Cook's distance (after DESeq)
dds <- DESeq(dds)
W <- results(dds)$cooksd
boxplot(W, main = "Cook's Distance")

# Identify outlier samples from PCA
pca <- prcomp(t(assay(vsd)))
outliers <- abs(scale(pca$x[,1])) > 3 | abs(scale(pca$x[,2])) > 3

from scipy import stats

z_scores = stats.zscore(pca_result, axis=0)
outliers = (np.abs(z_scores) > 3).any(axis=1)
print('Potential outliers:', counts.columns[outliers].tolist())

# Color PCA by batch
plotPCA(vsd, intgroup = c('condition', 'batch'))

# Test for batch effect
design(dds) <- ~ batch + condition
dds <- DESeq(dds)

# Color by batch in PCA
for batch in metadata['batch'].unique():
    mask = metadata['batch'] == batch
    plt.scatter(pca_result[mask, 0], pca_result[mask, 1],
                marker=['o', 's', '^'][list(metadata['batch'].unique()).index(batch)],
                label=f'Batch {batch}')

# Genes detected vs library size
plot(colSums(counts(dds)), colSums(counts(dds) > 0),
     xlab = 'Library Size', ylab = 'Genes Detected')

# Saturation check

plt.scatter(counts.sum(), (counts > 0).sum())
plt.xlabel('Total Counts')
plt.ylabel('Genes Detected')
plt.savefig('library_complexity.png')

# Most variable genes
rv <- rowVars(assay(vsd))
top_var <- order(rv, decreasing = TRUE)[1:500]

# Expression distribution
boxplot(log2(counts(dds) + 1), las = 2)

gene_var = log_cpm.var(axis=1).sort_values(ascending=False)
top_var_genes = gene_var.head(500).index

counts[top_var_genes].boxplot(figsize=(12, 6))
plt.xticks(rotation=45)
plt.savefig('gene_expression_dist.png')

# Quick summary
cat('Samples:', ncol(dds), '\n')
cat('Genes before filter:', nrow(counts), '\n')
cat('Genes after filter:', nrow(dds), '\n')
cat('Median library size:', median(colSums(counts(dds))), '\n')
cat('Median genes detected:', median(colSums(counts(dds) > 0)), '\n')

name	bio-rna-quantification-count-matrix-qc
description	Quality control and exploration of RNA-seq count matrices before differential expression. Check for outliers, batch effects, and sample relationships. Use when assessing count matrix quality before DE analysis.
tool_type	mixed
primary_tool	DESeq2

name	bio-rna-quantification-count-matrix-qc
description	Quality control and exploration of RNA-seq count matrices before differential expression. Check for outliers, batch effects, and sample relationships. Use when assessing count matrix quality before DE analysis.
tool_type	mixed
primary_tool	DESeq2

bio-rna-quantification-count-matrix-qc

خيارات التثبيت

مراجعة ملفات المصدر

Count Matrix QC

Load and Inspect Counts

R

Count Matrix QC

Load and Inspect Counts

R

Python

Basic Statistics

R

Python

Filter Low-Count Genes

R

Python

Normalize for Visualization

R (DESeq2 VST)

Python (log2 CPM)

Sample Correlation

R

Python

PCA Analysis

R

Python

Detect Outliers

R

Python

Check for Batch Effects

R

Python

Library Complexity

R

Python

Gene-Level QC

R

Python

Summary Report

Related Skills

bio-rna-quantification-count-matrix-qc

معلومات المصدر

خيارات التثبيت

مراجعة ملفات المصدر

المزيد من هذا المستودع

المزيد من هذا المستودع

Count Matrix QC

Load and Inspect Counts

R

Count Matrix QC

Load and Inspect Counts

R

Python

Basic Statistics

R

Python

Filter Low-Count Genes

R

Python

Normalize for Visualization

R (DESeq2 VST)

Python (log2 CPM)

Sample Correlation

R

Python

PCA Analysis

R

Python

Detect Outliers

R

Python

Check for Batch Effects

R

Python

Library Complexity

R

Python

Gene-Level QC

R

Python

Summary Report

Related Skills