Biomarker Discovery

GPU-accelerated analysis of gene-expression combinations to support biomarker discovery and patient stratification.

Business impact

• analysis runtime — The reported runtime dropped from 8.4 hours to 0.5 hours on the benchmark.
• computational speed — The article reports approximately 16.8 times faster analysis than traditional computing methods.
• gene combinations evaluated — The benchmark searched 7,331,162,245 possible three-gene combinations.

Data requirements

• clinical dataset (tabular) — The benchmark used a clinical dataset of 11 patients.
• gene-expression features (tabular) — The analysis used 3,531 gene-expression features.

AI methods and techniques

• proprietary algorithm — Used to search the gene-combination space and identify correlated triplets.
• GPU-accelerated quantum simulation — Used to execute the benchmark efficiently on GPU hardware.

Industries

Real-world evidence

1 documented case study on record.

Companies using this: Quantum X Labs.