ALE

[SanTanBan]

ALE: Autonomous Longevity Engine

ALE could be imagined as an intelligent AI, that is able to perform its own experiments focused on Human Longevity. Its driven by an LLM-AI connected to Databases and BioInformatics Tools. This will perform computations and simulations (leveraging the tools) to understand what experimentations could be designed for directing wet-lab research.

This will be implemented through the integration of AI-Agent(s) with Biological Databases and BioInformatics tools.

This multi-disciplinary issue has been also created in this URL: ZuGrama/longevity#6

[nooblemon-eth]

@SanTanBan could we please update this issue with an overview of the idea, a brief explanation of the problem statement and proposed solution?

This will help people understand if it's interesting enough to click out to the url to explore things in more detail. :)

[SanTanBan]

Reference: https://www.youtube.com/watch?v=l_OtimRlVAQ
Potential Competition: https://insilico.com/
Conversations: https://www.researchhub.com/question/3904/tools-to-connect-biological-pathways-to-existing-therapeutics
Join our Telegram Channel: https://t.me/+YSGdg4mZrmY2MmRl

[SanTanBan]

Please reach out to our current team and get involved in the development: https://github.com/orgs/ZuGrama/teams/openalers

Proposed pipeline:
MVP 1: OpenSource Deployment of AI Agents that can extract keywords from users' prompts, fetch research papers from ArXiv, and summarize them.
MVP 2: OpenSource Deployment of AI Agents that can generate detailed Methods and Procedures from a Biological HypoThesis.
MVP 3: Connect models with BioInformatic Tools and Datasets, for an enhanced proprietary version of our solution towards commercialization.

[SanTanBan]

MVP 1: https://research-synth-miliskrap.replit.app/
Resource 1) https://replit.com/@MiliSkrap/research-agents-zugrama

MVP 2: https://research-synthesizer.replit.app/
Resource 2) https://github.com/SanTanBan/ResearchSynthesizer

[SanTanBan]

Lite Paper: https://docs.google.com/document/d/1oDYqahKGelvr1idGGMeQp-DI4er9tmEXcXPgMc9Nj3U/

Presentation at Demo Day 1: https://docs.google.com/presentation/d/1zEVkw6_xwP_n8P-oLVFQ-rAuWMiKtjeMIOqNpKRwNrk/

Presentation at Demo Day 2: https://docs.google.com/presentation/d/1TLTQL5QdggnRIkdkyhg56Ee4n62wweOz/

Presentation at Demo Day 3: https://supersci.replit.app/

[SanTanBan]

Support our project here: https://antprotocol.vercel.app/posts/0x0172a17e03bf99f7a0293e063443dc9b97d78f28

Approximate Spending Plan for an estimated 8–10-month timeline ~>

Core Team Development: U$D 30,000/-
AI/ML Expert; Full Stack Developer; Computational Biology/Bioinformatics Expert

Stage 1: Autonomous Dry-Lab Scientist Development (Scigent)
AI Agentic-Crew Development and API Calls:
• Cloud Computing Infrastructure and AI Model API Costs: $30,000
• API calls for Bioinformatics tools (hosted at separate computation centres): $10,000
Compute Infrastructure:
• High-performance workstation with GPUs for development: $5000
Operational Costs:
• Co-working space (rental, 1 year), utilities and internet: $5,000
Prototype Development, Testing and Validation (R&D):
• Comprehensive Test of the AI Scientist leveraging research papers behind paywalls: $10,000

Stage 2: Integration and Network Building
Software Integration:
• Development of APIs for existing tools (for which APIs don’t exist): $10,000
• Integration with client softwares: $10,000
Network Development:
• Platform development for researcher collaboration, and for clarifying Agent queries: $10,000
• Outreach and partnership building: $10,000
Product Deployment:
• Development of client-facing platform: $2,500
• Deployment and Client Onboarding system development: $2,500

Intellectual Property Protection:
• Basic IP protection: $1,000
• Patent filings and legal consultations (Legal and Compliance): $1,000
Miscellaneous:
• Travel for conferences/networking: $2,500
• Contingency: $10,000

Total Budget: $150,000

[SanTanBan]

Here is the link to Research and Development/Funding Applications for developing ALE: https://youtube.com/playlist?list=PL4xbQJ0qLdqfAJWED7daNvJ5CVB8oKp1x