This course introduces participants to AI-based coding tools and demonstrates how to integrate them effectively into bioinformatics research workflows. The focus is on practical strategies to use AI for coding tasks in Python, including prompt engineering, quality assessment, reproducible documentation, and best practices for transparent, AI-assisted programming.

Participants will gain hands-on experience using AI tools to write, debug, and optimize Python scripts; evaluate AI-generated code; and apply these tools to genomics and data analysis challenges.

• Biologists or bioinformaticians who want to boost coding productivity using AI tools.

• Researchers interested in reproducibility, efficiency, and transparent documentation of AI-assisted workflows.

By the end of the course, participants will be able to:

• Write effective prompts to generate Python code using AI tools.

• Evaluate, debug, and improve AI-generated Python code for reliability and efficiency.

• Apply AI-assisted coding to data wrangling and analysis tasks in bioinformatics.

• Document AI-assisted workflows for reproducibility and transparency.

• Confidently integrate AI tools into everyday Python-based bioinformatics workflows.

Day 1 – Getting Started with AI-assisted Coding - 2-6 PM Berlin time

• Introduction to AI, large language models, and AI coding assistants: strengths, limitations, and examples of good and bad AI-generated code.

• Prompt Engineering 101: how to design effective prompts for Python coding tasks.

• Tools overview: working with Jupyter notebooks: integrating AI into Python projects (scripts vs. notebooks).

• Hands-on: generating Python scripts with AI for simple tasks (data parsing, file I/O, data manipulation with pandas).

• Best practices for coding with AI: reproducibility, transparency, and ethical use.

• Evaluating AI-generated code: correctness, efficiency, readability.

• Debugging patterns: handling AI “hallucinations,” verifying functions, and checking documentation.

• Hands-on: debugging and improving AI-generated Python scripts.

• Mini-project: writing prompts to generate a script for basic data wrangling in genomics.

Day 2 – Advanced Workflows & Reproducibility -- 2-6 PM Berlin time

• Reproducibility essentials: seeds, data snapshots, prompt journals, and documentation templates.

• Advanced prompting strategies: iterative prompting, role prompting, and handling errors in AI-generated Python code.

• Hands-on: using AI to automate repetitive coding tasks (loops, pipelines, plot generation with matplotlib / seaborn).

• Quality assurance: testing, validation, and comparing AI-generated code solutions.

• Documenting AI-assisted workflows: notebooks, comments, metadata, and session info.

• Applied example: AI-assisted analysis of genomic or tabular data in Python – from raw input to tidy table to QC plots (e.g., pandas, matplotlib).

• Hands-on project: design a small reproducible Python script with AI assistance using a provided dataset, document it, and present results (lightning presentations).

• Wrap-up discussion: the future of AI in coding and bioinformatics – open Q&A.

1 - Introduction to Python for biologists - ONLINE, 9-12 February

2 - Advanced Python for Data Science and Bioinformatics - ONLINE, 23-26 March

3 - AI for Genomics - ONLINE, 7-9 April

4 - Genomic Data Visualisation with Python - ONLINE, 22-24 June

Should you have any further questions, please send an email to [email protected]