• Register
X
Forgot Password

If you have forgotten your password you can enter your email here and get a temporary password sent to your email.

X

Leaving Community

Are you sure you want to leave this community? Leaving the community will revoke any permissions you have been granted in this community.

No
Yes
X
Forgot Password

If you have forgotten your password you can enter your email here and get a temporary password sent to your email.

dkNET Webinar: Integrating RAG Chatbots into Bioinformatics Platforms


Join dkNET Webinar on Friday, May 16, 2025, 11 am - 12 pm PT


Presenter: Avi Ma'ayan, PhD, Mount Sinai Endowed Professor in Bioinformatics, Professor in Department of Pharmacological Sciences, and Director of Mount Sinai Center of Bioinformatics, Icahn School of Medicine at Mount Sinai


Abstract

The advent of large language models (LLMs) has revolutionized numerous fields with their impressive performance on a variety of natural language processing tasks. However, their application in biomedical research remains controversial due to their well-documented tendency to generate non-factual content. Given the foundational importance of accuracy, rigor, and reproducibility in the life sciences, such limitations pose significant barriers to adoption. Retrieval-Augmented Generation (RAG) frameworks offer a promising solution by grounding LLM responses with curated external knowledge sources, such as databases or document repositories, thereby constraining outputs to verifiable and citable information. Over the past two years, the Ma’ayan Lab has integrated RAG-based chatbots into several of its web-based bioinformatics platforms to enhance user interaction and provide accurate, context-specific assistance. These platforms include Playbook Workflow BuilderD2H2GeneSetCartHarmonizome, and the CFDE Workbench. In this presentation, I will provide an overview of each of these software platforms, highlighting how RAG-enhanced conversational interfaces were implemented to improve usability, facilitate knowledge discovery, and support data-driven discovery with examples related to diabetes.

Links to RAG chatbot features in software platforms developed by the Ma'ayan Lab:
Diabetes Data and Hypothesis Hub (D2H2): https://d2h2.maayanlab.cloud/hypotheses


The top 3 key questions that Diabetes Data and Hypothesis Hub(D2H2) resource can answer:

1. What transcription factors or small molecules are associated with my gene set in the context of diabetes? 

2. Which diabetes-related studies show significant differential expression of my gene or gene set?

3. Which LINCS L1000 small molecules may reverser or mimic my up and down gene sets?


Dial-in Information: https://uchealth.zoom.us/meeting/register/91I6Og3PQwCIJe_iXBLjPQ

Date/Time: Friday, May 16, 2025, 11 am - 12 pm PT


Upcoming webinars schedule: https://dknet.org/about/webinar


X

Are you sure you want to delete that component?