US 2025/0299779 A1
Published U.S. patent application
Deep Learning-Based System for Rapid and Accurate Bacterial Classification
A system for classifying bacterial genomic sequences using convolutional and recurrent neural-network models.
Inventorship · Galacto Corp.
Published patent applications
Daniel J. Mueller is the sole named inventor on two published U.S. patent applications assigned to Galacto Corp.
- 2 published U.S. applications
- Sole named inventor
- Galacto Corp.
- AI and bioinformatics
- Semantic compression
Patent record
Two published applications
Each record links to Justia and the official USPTO publication.
US 2026/0220089 A1
Published U.S. patent application
Semantically Preserved File Compression System Utilizing Optimized Lookup Tables Enabling Active Processing of Compressed Data
A multi-level compression system and file format using optimized lookup tables to preserve semantic content and support processing of compressed data.
01
From algorithm research to published applications
The two applications cover distinct but related areas of computational research. One addresses bacterial classification from genomic sequences using deep-learning architectures. The other describes a multi-level compression system designed to preserve semantic content while supporting operations on compressed data.
Both records name Daniel J. Mueller as the sole inventor and Galacto Corp. as applicant and assignee. Each card links to its Justia record and the official U.S. Patent and Trademark Office publication PDF.
02
Inventorship in the public record
Publication places each application in the public record with Daniel’s name, filing details, abstract, specification, and claims. The records connect his work at Galacto Corp. with two original technical domains.
The applications are presented by their precise publication numbers and titles so the work can be found, cited, and reviewed directly.
03
Technical scope
The bacterial-classification application combines genomic preprocessing with convolutional and recurrent neural-network approaches for genus and species classification. The compression application describes hierarchical characterization, optimized lookup tables, a dedicated file format, and access through a service interface.
The linked USPTO publications provide the complete public specifications and claims.