Guilherme Medeiros Martins

Title: Guilherme Medeiros Martins: Inventor in Machine Learning-Based Genotyping

Introduction

Guilherme Medeiros Martins is an inventor based in San Diego, California, and is associated with Illumina Corporation. His work includes a patented approach that applies machine learning to predict outcomes in genotyping processes.

Latest Patents

Martins is named on one patent, titled “Machine learning-based genotyping process outcome prediction using aggregate metrics.” The disclosed method analyzes images of beads arranged in tiles on a beadchip. It calculates average full width at half maximum (FWHM) values for bead images across image channels, then uses those aggregate measurements to predict the likelihood that a sample evaluation run will fail. A trained classifier may generate the failure-likelihood score, which the system reports to help avoid unnecessary post-processing.

Career Highlights

Martins’s patent reflects an application of machine learning to sample evaluation and image-based genotyping workflows. By predicting potential run failures from aggregate image metrics, the approach is designed to support more efficient processing.

Collaborations

The provided information identifies Martin Detelinov Martchovsky and Naghmeh Rezaei as Martins’s coworkers at Illumina Corporation.

Conclusion

Guilherme Medeiros Martins’s patented work combines beadchip image analysis and machine learning to predict genotyping process outcomes and help streamline sample evaluation.