Quantum X Labs reports its GPU-accelerated platform achieved approximately 16.8 times faster analysis of gene combinations than traditional computing methods. The company’s algorithm searched 7,331,162,245 possible three-gene combinations within a dataset of 3,531 genes from 11 patients, identifying 22.2 million correlated gene triplets. Analysis that previously required 8.4 hours using 64 vCPUs now completes in just 0.5 hours with a single NVIDIA T4 GPU. “This represents an important development milestone for Quantum X Labs,” said Dr. Tidhar Turgeman, head of clinical trials data analysis, highlighting progress toward scalable biomarker discovery.

GPU Acceleration Achieves 16.8x Speedup in Gene Combination Analysis

The analysis of over seven billion three-gene combinations was completed in 0.5 hours using a single NVIDIA T4 GPU, a process that previously required 8.4 hours with 64 vCPUs. Quantum X Labs achieved this acceleration while analyzing a clinical dataset of 11 patients and 3,531 gene-expression features, demonstrating performance gains in a high-dimensional biological setting. The company’s proprietary algorithm, operated through its subsidiary CliniQuantum, identified approximately 22.2 million correlated gene triplets using a predefined correlation threshold.

This benchmark, conducted on an AWS g4dn.16xlarge EC2 machine, used IBM Qiskit Aer GPU-based quantum simulation to evaluate computational performance. The substantial speedup stems from a GPU-accelerated quantum-simulation platform, achieving a 16.8-fold increase in computational speed compared to CPU-only implementation on the same hardware.

This advance addresses a key challenge in clinical data analysis, where the number of variables often far exceeds the number of patients, creating vast combinatorial search spaces. Dr. Turgeman explained, “Clinical and molecular datasets can contain thousands of variables but relatively few patients, creating enormous combinatorial search spaces.” Demonstrating the ability to evaluate more than 7.3 billion potential gene combinations on real clinical data marks a step in our development.

The benchmark’s design focused solely on computational performance and did not assess the clinical validity or predictive value of the identified gene combinations. Quantum X Labs intends to extend this work to analyze even more complex multi-feature configurations and larger biomedical datasets. The company is also progressing toward executing these algorithms on actual quantum computing hardware, building on the current simulation and validation phase.

Quantum X Labs Ltd. focuses on developing quantum algorithms for applications including transportation, drug discovery and security, alongside quantum-based GPS replacement and atom accuracy solutions. Gix Media and Metagramm, also subsidiaries, develop software solutions for digital advertising and grammatical error correction, respectively, using artificial intelligence and machine learning technologies. Tidhar Turgeman leads the company’s work, which aims to enable scalable biomarker discovery and patient stratification through quantum-enabled approaches.