Genovate

Why I Built It

I built GENOVATE to create a practical bridge between what I was learning about CRISPR/Cas9, genetics, kidney disease, and computational biology and one of the most important challenges in gene therapy: deciding how gene-editing systems should actually be delivered.

The original motivation came from my grandfather’s kidney condition and my interest in Polycystic Kidney Disease (PKD). While researching mutations in PKD1 and PKD2, I realized that identifying a disease-causing gene or designing a guide RNA is only part of the problem. A gene-editing system still has to reach the correct cells safely, efficiently, and with minimal off-target effects.

That led me to focus on two very different CRISPR-delivery strategies: electroporation and lipid nanoparticles (LNPs). I wanted to build a platform that could help users compare these approaches computationally before moving into costly laboratory experimentation.

GENOVATE began as a PKD-focused CRISPR delivery simulator, but gradually expanded into a broader computational-biology and educational platform incorporating gene analysis, mutation mapping, sequence exploration, CRISPR target identification, delivery comparison, and learning resources.

The broader goal is to make advanced genomics and gene-editing concepts more accessible to students and early-stage researchers, particularly those who may not have access to research-grade genomics infrastructure. I also see GENOVATE as a foundation for a future biotechnology platform capable of supporting more advanced research and clinical-research workflows.

How It Works

GENOVATE takes biological and experimental inputs from the user and uses them to analyze CRISPR targets and compare possible delivery strategies.

Users can provide information such as target organ, target gene or mutation, PKD-related mutations including PKD1, PKD2, or PKHD1, PAM sequence, an optional FASTA DNA sequence, and experimental or delivery parameters used within the simulation.

For sequence-based workflows, GENOVATE can analyze DNA, identify relevant genomic information, scan for NGG PAM sites, extract possible 20-base SpCas9 target sequences, and connect sequence or mutation information to downstream CRISPR-delivery analysis.

The platform then compares lipid nanoparticles and electroporation using factors such as editing efficiency, off-target probability, transfection success, cell viability, tissue-targeting effectiveness, scalability, cost, precision, biosafety, and translational feasibility.

Based on these parameters, GENOVATE generates a recommended delivery strategy, comparative performance scores, and confidence-style outputs. For example, in a kidney-focused PKD1 scenario, the system can evaluate whether LNP delivery or electroporation appears more suitable under the selected conditions.

The core workflow is: Input biological target → identify or match gene/mutation → analyze CRISPR target → compare delivery methods → score performance → recommend strategy → visualize results → export research plan/report.

GENOVATE is designed as an educational and decision-support platform, not as a clinically validated treatment-selection system.

Features

CRISPR Delivery Simulation — GENOVATE compares LNPs and electroporation across multiple biological and technical parameters, including editing efficiency, off-target risk, transfection success, cell viability, tissue targeting, scalability, and cost. The platform can recommend a preferred strategy and provide comparative or confidence-style outputs while allowing users to adjust parameters and explore different scenarios.

Gene Detection — Sequence-based workflows can be used to identify or match genes using NCBI BLAST-informed analysis, helping connect an input DNA sequence with relevant genomic information.

Mutation Mapping — GENOVATE supports mutation-focused analysis, particularly for PKD1, PKD2, and PKHD1, using NCBI and Entrez-informed workflows to connect mutation information with CRISPR-targeting and delivery analysis.

Sequence Viewer — The platform can display genomic or transcript sequence information, support sequence inspection, visualize PAM locations, and highlight regions relevant to CRISPR targeting.

PAM and gRNA Analysis — GENOVATE can scan sequences for NGG PAM sites, identify candidate 20-bp SpCas9 target sequences, and support guide-target analysis using on-target and off-target scoring concepts.

Comparative Visualizations — Results can be presented through radar charts, bar charts, comparative scores, and visual performance summaries, helping users interpret complex delivery trade-offs more clearly.

Personalized Editing Plan — Based on the selected biological target and delivery parameters, the system can generate a structured editing recommendation explaining the preferred delivery method and the reasoning behind it.

Downloadable Reports — GENOVATE can generate PDF research plans and analytical reports, with some versions also supporting CSV-style data export.

Learning Mode — A dedicated educational component explains concepts related to CRISPR/Cas9, lipid nanoparticles, electroporation, mutations, and gene editing. It includes glossary-style explanations and external learning resources so that students can use the platform even without advanced prior knowledge.

Family / Inheritance Tools — A pedigree-style family-chart feature was explored to help users visualize inheritance patterns and genetic risk, using graph-based representations.

Collaborative Research — Later GENOVATE concepts expanded toward allowing users to save projects, simulation runs, notes, and comments, with the aim of supporting collaborative work among students and early-stage researchers.

The platform has been envisioned for potential use by students, universities, IB schools, and early-stage clinical researchers.

Technology

GENOVATE was originally developed primarily using Python and Streamlit, combining bioinformatics, machine learning, visualization, genomics databases, and web development within a single platform.

The computational and scientific stack has included Python — core programming language, Streamlit — web application framework, pandas — data handling and structured datasets, NumPy — numerical operations, scikit-learn — machine-learning workflows, Random Forest models — delivery recommendation and classification, matplotlib — data visualization, FPDF — downloadable PDF report generation, Biopython — sequence handling and biological data processing, NCBI BLAST — gene and sequence identification, NCBI Entrez — genomic and gene-information retrieval, GenBank / NCBI Gene — sequence and gene data, Ensembl — genomic and mutation context, PubChem — molecular and chemical information used in broader research workflows, and Graphviz — pedigree and family-chart visualization.

The platform also drew inspiration from professional genomics and CRISPR tools such as CRISPOR, UCSC Genome Browser, CHOPCHOP, and Benchling, particularly in areas such as guide-RNA selection, PAM visualization, and off-target analysis.

Earlier versions used a Streamlit-based architecture, while later expansion plans explored a more scalable structure using a React frontend, FastAPI backend, and Supabase for user accounts, projects, and data storage.

Depending on the module, GENOVATE uses either a machine-learning model or a weighted scoring framework to compare delivery approaches. Its recommendations are based on literature-informed CRISPR-delivery parameters and computational reasoning rather than direct clinical validation.

Overall, GENOVATE combines bioinformatics, machine learning, CRISPR sequence analysis, genomics databases, scientific visualization, and web development into a single platform designed to make gene-editing delivery analysis more accessible and understandable.