Awesome

awesome-compbio-chatgpt

An awesome repository of community-curated applications of ChatGPT and other LLMs in computational biology!

Any material is welcome, as long as it is:

• Free to read without accounts
• High quality (subjective, but we know it when we see it)
• Relevant to bioinformaticians

To contribute, just add the link to this document and open a Pull Request!

Have fun!

Prompt collections

• Data science prompts: https://github.com/travistangvh/ChatGPT-Data-Science-Prompts
• ChatGPT cheatsheet for Data Science: https://www.datacamp.com/cheat-sheet/chatgpt-cheat-sheet-data-science
• Fun and useful prompts: https://github.com/f/awesome-chatgpt-prompts
• List of awesome ChatGPT lists: https://github.com/OpenMindClub/awesome-chatgpt
• Collection of prompts for developers (YouTube): https://www.youtube.com/watch?v=sTeoEFzVNSc&t=901s

Applications that use the GPT API

• GPT for Google Sheets: https://gptforwork.com/
• GPT for R Studio: https://github.com/MichelNivard/gptstudio
• GPT for R Developers: https://jameshwade.github.io/gpttools/
• GPT for PDFs (in a web interface, freemium service): https://chatpdf.com
• GPT directly from the command line: https://github.com/npiv/chatblade
• GPT-based chatbot fine-tuned for bioinformatics https://ai.tinybio.cloud/chat (discussed on Biostars)
• Very powerful question answering using GPT and internet searches: https://www.perplexity.ai/
• Python framework for connecting conversational AI to bioinformatics pipelines https://github.com/biocypher/biochatter and ChatGSE

Prompt Engineering

• Quick guide of best practices for Prompt Engineering: https://help.openai.com/en/articles/6654000-best-practices-for-prompt-engineering-with-openai-api
• More advanced guidance on Prompt Engineering: https://www.promptingguide.ai/
• Good 20 min blogpost covering core aspects of Prompt Engineering: https://lilianweng.github.io/posts/2023-03-15-prompt-engineering/

Understanding the tool

• A very good post by Stephen Wolfram on the details on how it works: https://writings.stephenwolfram.com/2023/02/what-is-chatgpt-doing-and-why-does-it-work/
• List of papers published by OpenAI: https://openai.com/research
• Basic YouTube Crash Course on ChatGPT (12/December/2022, so a bit old): https://www.youtube.com/watch?v=JTxsNm9IdYU
• Technnical overview of details on how LLMs work and what can be their future (Eight Things to Know about Large Language Models): https://cims.nyu.edu/~sbowman/eightthings.pdf

Ethics and Accountability Discussions

• "Pause Giant AI Experiments: An Open Letter" https://futureoflife.org/open-letter/pause-giant-ai-experiments/

Bioinformatics-specific resources

• Ten Quick Tips for Harnessing the Power of ChatGPT/GPT-4 in Computational Biology (our article): https://arxiv.org/abs/2303.16429 and https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1011319

• Single-cell RNA-seq analysis in Python guided by ChatGPT (nice video, from Cell Ranger output to final plots): https://www.youtube.com/watch?v=fkuLFlC2ZWk

• Using ChatGPT in bioinformatics and biomedical research (good blog post): https://omicstutorials.com/using-chatgpt-in-bioinformatics-and-biomedical-research/

• ChatGPT for bioinformatics (on the perspective of a bioinformatician): https://medium.com/@91mattmoore/chatgpt-for-bioinformatics-404c6d0817a1

• ChatGPT and bioinformatics careers (Reddit forum discussion): https://www.reddit.com/r/bioinformatics/comments/11wwnqj/chatgpt_and_bioinformatics_careers/

• ChatG-PPi-T: Finding Interactions with OpenAI (far from production-ready, but creative idea): https://www.linkedin.com/pulse/chatg-ppi-t-finding-interactions-openai-jon-hill/

• BioGPT: generative pre-trained transformer for biomedical text generation and mining: https://github.com/microsoft/BioGPT

• A Platform for the Biomedical Application of Large Language Models: https://arxiv.org/abs/2305.06488

• scGPT: Foundation model for single-cell biology: https://github.com/bowang-lab/scGPT (Paper (not OA))

API and advanced applications

• LangChain combines LLM/GPT API requests with (1) access to external documents and (2) abilities to talk to the wider web, creating semi-autonomous agents: https://python.langchain.com/en/latest/
• Llama Index is an interface for combining LLMs with external data (such as documents): https://gpt-index.readthedocs.io/en/latest/index.html
• guardrails is an application to improve and refine LLM outputs in Python: https://github.com/ShreyaR/guardrails
• Embedchain is a framework to create ChatGPT like bots over your personalised dataset in 3 lines of code: https://embedchain.ai/
• The Hugging Face Hub open LLM leaderboard collects and ranks open-source LLMs; very important for accessibility and reproducibility of LLM-based bioinformatics

Quick Prompts

The following prompts may not be the most efficient in terms of prompt engineering, but they are quick, useful and exemplify usage of ChatGPT for computational biologists.

Improve Code Readability and Documentation

• “Add explanatory comments to this code: {code here}”
• “Rename the variables for clarity: {code here}”
• "Render roxygen2 documentation for the function: {R code here}”.

Write Code Efficiently

• "Extract functions to increase modularity: {code here}"
• "Write a unit test for the following function and help me implement it: {code here}"
• "Re-write and optimize this for loop: {code here}"

Clean up data

• "Write me regex for R/python/Excel with a pattern that will extract {} from {}"
• "Act as a table. Add a new column with consistent labels to this dataset:"

Improve data visualization

• "Create a ggplot2 violin plot with a log10 Y axis"
• "Change my code to make the plot color-blind friendly"
• "Translate this plot from ggplot2 to matplotlib syntax"

Tutorials

• "A Guide to Using ChatGPT For Data Science Projects" : https://www.datacamp.com/tutorial/chatgpt-data-science-projects