Working with Packages

Python comes with a standard library of built-in modules. These cover common tasks like file handling, math, and text processing. For specialized work like data analysis or bioinformatics, you will need third-party packages from PyPI. PyPI hosts over 500,000 packages that extend what Python can do.

Importing packages

Before using a package, you need to import it. Python offers several ways to do this.

import

The simplest approach imports the entire module.

import os

print(os.getcwd())

/home/user/projects/rna_seq_analysis

You then access functions and classes using the module name as a prefix.

from … import

You can import specific items from a module. This lets you use them directly without the module prefix.

from pathlib import Path

p = Path("sample_data.fastq.gz")
print(p.stem)
print(p.suffix)
print(p.name)

sample_data.fastq
.gz
sample_data.fastq.gz

Path is very useful for working with file paths in bioinformatics pipelines. It handles path separators across operating systems automatically.

Aliases

Some packages have long names. The community has adopted standard short aliases for common packages. Use the as keyword to create an alias.

import numpy as np

counts = np.array([100, 250, 50, 300, 75])
print(counts)
print(counts.mean())
print(counts.std())

[100 250  50 300  75]
155.0
100.4987562112089

You will see np for NumPy, pd for pandas, and plt for matplotlib in nearly every tutorial and codebase. Stick with these conventions so your code is readable to others.

Standard library highlights

The standard library includes many useful modules. Here are a few that come up often in bioinformatics work.

Module	Purpose
os	Interact with the operating system, environment variables
pathlib	Object-oriented file path handling
csv	Read and write CSV/TSV files
collections	Specialized container types like Counter and defaultdict
gzip	Read and write gzip-compressed files
json	Parse and write JSON data
sys	Access command-line arguments and system info

collections.Counter

Counter tallies how often each item appears in a list. This is handy for counting gene occurrences, k-mers, or any repeated elements.

from collections import Counter

gene_list = ["TP53", "BRCA1", "TP53", "EGFR", "BRCA1", "TP53"]
gene_counts = Counter(gene_list)
print(gene_counts)
print(gene_counts.most_common(2))

Counter({'TP53': 3, 'BRCA1': 2, 'EGFR': 1})
[('TP53', 3), ('BRCA1', 2)]

csv module

Many bioinformatics tools produce tab-separated output files. The csv module handles both CSV and TSV formats.

import csv
import io

tsv_data = "gene\tlog2fc\tpvalue\nTP53\t2.5\t0.001\nBRCA1\t-1.8\t0.003\nEGFR\t3.2\t0.0001"
reader = csv.DictReader(io.StringIO(tsv_data), delimiter="\t")
for row in reader:
    print(f"{row['gene']}: log2fc={row['log2fc']}, p={row['pvalue']}")

TP53: log2fc=2.5, p=0.001
BRCA1: log2fc=-1.8, p=0.003
EGFR: log2fc=3.2, p=0.0001

DictReader maps each row to a dictionary using the header as keys. This makes your code more readable than indexing by column number.

Installing packages with uv

This project uses uv for Python package management. Do not use pip install directly. uv is faster, more reliable, and handles virtual environments automatically.

Setting up a new project

First, create a project directory and initialize it with uv.

mkdir gene_analysis
cd gene_analysis
uv init

This creates a pyproject.toml file and a .venv/ virtual environment.

Adding packages

Use uv add to install packages. This updates your pyproject.toml automatically.

uv add pandas numpy

You can add multiple packages at once. Each package and its version will be recorded in pyproject.toml.

Running scripts

Use uv run to execute scripts in the correct environment.

uv run python script.py

This ensures Python uses the packages installed in your project’s virtual environment.

Never use pip install directly

Always use uv add to install packages. Running pip install bypasses the project’s dependency tracking and can cause conflicts. If you see instructions elsewhere that say pip install, replace them with uv add.

pyproject.toml

After running uv init and adding some packages, your pyproject.toml will look something like this:

[project]
name = "gene-analysis"
version = "0.1.0"
description = ""
readme = "README.md"
requires-python = ">=3.12"
dependencies = [
    "numpy>=2.2.4",
    "pandas>=2.2.3",
]

This file is the single source of truth for your project’s dependencies. When you share your project, others can run uv sync to install all the same packages.

Key bioinformatics packages

These are the most commonly used third-party packages in bioinformatics and data science.

Package	Purpose	Install command
pandas	Tabular data manipulation	uv add pandas
polars	Fast DataFrame library for large datasets	uv add polars
numpy	Numerical arrays and math operations	uv add numpy
scipy	Statistical tests and scientific computing	uv add scipy
scikit-learn	Machine learning algorithms	uv add scikit-learn
scanpy	Single-cell RNA-seq analysis	uv add scanpy
pysam	Read and write BAM/SAM/VCF files	uv add pysam
biopython	Sequence analysis, BLAST parsing, file I/O	uv add biopython
matplotlib	General-purpose plotting	uv add matplotlib
seaborn	Statistical data visualization	uv add seaborn

You do not need all of these at once. Add packages as your project requires them.

Virtual environments

A virtual environment is an isolated Python installation for a single project. It prevents packages from one project from interfering with another.

When you run uv init, it creates a .venv/ directory inside your project folder. This directory contains the Python interpreter and all installed packages for that project.

You do not need to activate the virtual environment manually. Running uv run python script.py handles this for you. If you want to activate it yourself for interactive use, run:

source .venv/bin/activate

Each project should have its own virtual environment. Never install packages into your system Python.

Quick reference

Task	Command
Create a new project	uv init
Add a package	uv add <package>
Add multiple packages	uv add pandas numpy scipy
Remove a package	uv remove <package>
Run a script	uv run python script.py
Sync dependencies from pyproject.toml	uv sync
Import a module	import module_name
Import a specific item	from module import item
Import with alias	import numpy as np

Next steps

Now that you know how to install and import packages, you are ready to work with tabular data. Head to Pandas Essentials to learn how to load, filter, and analyze data with pandas.