# CLEdgeSeq: Advanced sequence analysis and communication for edge applications > **NIH NIH R43** · QUANTITATIVE BIOSCIENCES, INC. · 2024 · $350,000 ## Abstract Developing effective and safe therapeutics is critical to our quality of life, and novel approaches are needed to improve the speed and accuracy of detection biological contamination and other quality control parameters. Edge DNA and RNA sequence computational capability will provide point-of-production and point-of-care support for analysis of biological contamination that will be essential for effective mon- itoring and gating of biological contaminants in water and therapeutic supplies. The main objective of this SBIR is to create a neural network-accelerated DNA and RNA sequence processing node suitable for field, clinical, and industrial use, making it of great utility to monitoring efforts aimed at tracking and assessing potential haz- ardous exposures, as well as highly marketable to a wide range of end-users. In order to demonstrate technical feasibility, QBI will perform these Specific Aims: Specific Aim 1: Develop neural network NGS data processing routines for mapping, assembly, and analysis in edge applications. DNA and RNA sequence analysis, which is core to a number of biologi- cal contaminant monitoring techniques but is also core to QBI's recent nanobody library technology, has traditionally has been reserved for servers and other capable computers, but neural network approaches have enabled modern edge computational platforms to perform sophisticated analyses outside of central servers. We seek to develop two proof of principle neural networks that accelerate two sequence analysis routines central to sequence analysis on edge platforms, which will permit distributed and cost-effective execution of sequence analysis in remote contexts. The error tolerance of neural networks will be espe- cially useful, since nanopore sequencers with relatively high sequence error rate are becoming a common edge technology due to their ability to sequence samples with minimal hardware and with minimal pre- processing of sample. Specific Aim 2: Develop and demonstrate CLEdgeSeq node as a solution for distributed NGS data processing. These neural network tools will be combined with data format specifications and network communication routines to establish feasibility of a computational node, CLEdgeSeq. CLEdgeSeq will pre-process NGS data at the point of sequencing, perform necessary analyses, assemble consensus se- quences, and then relay the results to a central server in a bandwidth-effective manner. The resulting platform will be a distributed computing approach to NGS data activities that should scale well with the expected rapid rise of NGS data generation from point-of-production and point-of-care sequencers, especially including nanopore sequencers. Successful completion of these Aims will serve to validate modern neural network approaches to sequence anal- ysis for edge hardware, making it of great utility to monitoring efforts aimed at tracking and assessing potential hazardous exposures. Creating this computational capability will be transforma... ## Key facts - **NIH application ID:** 11007332 - **Project number:** 1R43GM156154-01 - **Recipient organization:** QUANTITATIVE BIOSCIENCES, INC. - **Principal Investigator:** William Mather - **Activity code:** R43 (R01, R21, SBIR, etc.) - **Funding institute:** NIH - **Fiscal year:** 2024 - **Award amount:** $350,000 - **Award type:** 1 - **Project period:** 2024-08-01 → 2025-07-31 ## Primary source NIH RePORTER: https://reporter.nih.gov/project-details/11007332 ## Citation > US National Institutes of Health, RePORTER application 11007332, CLEdgeSeq: Advanced sequence analysis and communication for edge applications (1R43GM156154-01). Retrieved via AI Analytics 2026-05-26 from https://api.ai-analytics.org/grant/nih/11007332. Licensed CC0. --- *[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*