RNA sequencing (RNA-seq) is a powerful technique used to measure gene expression levels across the entire transcriptome. In a typical RNA-seq experiment, RNA molecules are extracted from biological samples, converted into complementary DNA (cDNA), and then sequenced. The resulting sequencing reads are mapped back to a reference genome or transcriptome, allowing you to determine which genes are being expressed and at what levels. The core data produced by this process is a gene count table, where each entry represents the number of sequencing reads that have been assigned to a particular gene in a particular sample.

Key steps in RNA-seq experiments:

• Extract RNA from biological samples;
• Convert RNA to cDNA;
• Sequence the cDNA to generate reads;
• Map sequencing reads to a reference genome or transcriptome;
• Generate a gene count table with read counts for each gene and sample.

A gene count table provides the foundation for downstream analysis of gene expression, enabling you to quantify and compare gene activity across different samples or experimental conditions.

# Example of a gene count table (not runnable code, for illustration only):

| Gene_ID   | Sample_1 | Sample_2 | Sample_3 |
|-----------|----------|----------|----------|
| ACTB      | 10500    | 9800     | 11200    |
| GAPDH     | 8700     | 9100     | 8800     |
| TP53      | 420      | 390      | 510      |
| MYC       | 2200     | 2400     | 2100     |
| EGFR      | 1150     | 1230     | 1190     |

The numbers in a gene count table represent read counts—the total number of sequencing reads that have been mapped to each gene for each sample. These counts provide a quantitative measure of gene expression: higher counts generally indicate higher expression of that gene in the sample. By comparing these counts across multiple samples, you can identify genes that are differentially expressed under different conditions or between different groups.

To analyze RNA-seq count data in Python, you will often start by loading the count table into a pandas DataFrame. Summing the counts for each sample helps you understand the library size—the total number of reads sequenced for each sample. This information is important for downstream analysis and normalization.

import pandas as pd

# Read a count table from a CSV file
# The CSV should have genes as rows and samples as columns
count_table = pd.read_csv("gene_counts.csv", index_col=0)

# Summarize total counts per sample (library size)
library_sizes = count_table.sum(axis=0)
print("Total counts per sample (library size):")
print(library_sizes)

While RNA-seq count data is a valuable measure of gene expression, you need to recognize common sources of bias:

• Library size differences can occur if some samples are sequenced more deeply than others, leading to higher counts that do not necessarily reflect true biological differences;
• Gene length can affect counts, as longer genes are more likely to capture more reads simply because they cover a larger region.

To account for these and other biases, you use normalization methods. Normalization adjusts raw counts so that gene expression measurements are more comparable across samples and genes, making your analysis more accurate and reliable.