Tools & Chat on Bio.Dev
Date Published
TLDR: I’ve tried to curate some of the most popular biotech tools, services, libraries, and ML models and integrate them with a chatbot. This is isn’t an exhaustive list, but aims to be a “first stop” for students and professionals who want to explore the landscape.
The ecosystem of software tools for biology is vast. There are thousands of tools, libraries, SaaS services that solve different problems. More recently, there has been an explosion of ML models for everything from protein design to molecular dynamics. This is incredible to see.
One challenge that remains is discoverability, especially for students and those new to the industry who might not know where to begin. Much of the information for biotech tools is fragmented across various sources including GitHub repos, scientific publications, PiPy docs, and lab or organization websites.
ChatGPT helps tremendously if you’re just starting a project and need some ideas for where to look, but I think having a web page to interactively browse through the landscape provides a better experience for exploration.
There have been some really cool efforts in the community to create something of this sort. These include the Awesome Bioinformatics repo by Daniel Cook, and the The Life Science Software Landscape by DeepOrigin. Both are incredible resources and you should check them out. I wanted to build upon these efforts on Bio.Dev to create a living and evolving catalog that sits alongside a blog with deep dive articles into specific tools and technologies.
Another neat feature which I thought would be cool is a chatbot that can live on the same site and be integrated with the catalog. This way you can reference specific tools (by using the @ followed by tool name) to automatically provide the bot additional context that it may not otherwise have. If in the course of your conversations the bot itself brings up a tool, it’ll automatically be rendered as a card in the chat window with a link to the corresponding Bio.Dev catalog item. You can then click on this to find additional information and links to that tool’s GitHub, DockerHub, docs, PyPi, scientific publication, and other relevant resources.
The catalog itself is very much a work in progress. I used a combination of previously published resources, an AI agent, and my own previous experience to try to aggregate an initial selection of tools. I also used an extract of the web contents from the URLs available from each tool to generate summaries, though these do leave something to be desired.
I will keep improving the quality of the catalog so that it’s expansive and informative, but not overwhelming, prioritizing those tools which are well documented and/or widely used. If you know of any tools which I’ve missed, please do go ahead and send them my way and I’ll be happy to add them!