Google’s Deep Somatic AI Tool Enhances Cancer Mutation Detection
Google has announced the launch of its new AI tool, “Deep Somatic,” which aims to improve the accuracy of identifying cancer-related genetic mutations in tumor genetic sequences. This achievement marks an important step towards more effective cancer treatment by providing a deeper understanding of the genetic mutations driving tumor growth.
The Importance of Identifying Genetic Mutations in Cancer Treatment
Cancer begins when there is a malfunction in the mechanisms controlling cell division. Identifying specific genetic mutations that lead to tumor growth is crucial for developing effective treatment plans. Doctors regularly sequence the genomes of tumor cells from biopsies to guide treatments targeting the growth and spread of specific cancers.
The work related to the Deep Somatic tool has been published in Nature Biotechnology. It uses convolutional neural networks to identify genetic variants in tumor cells with greater accuracy than current methods. Google has made the tool and its high-quality training data openly available.
The Challenge of Somatic Variants in Cancer Genetics
Cancer genetics is highly complex. While genome sequencing reveals cancer-associated genetic changes, distinguishing between true variants and sequencing errors is a significant challenge, where AI tools can provide the needed assistance. Most cancers are attributed to somatic variants acquired after birth, rather than inherited genetic variants from parents.
Somatic mutations occur when DNA is exposed to environmental factors like UV radiation or when random errors occur during DNA replication. When these variants alter normal cell behavior, they can cause uncontrolled division, leading to cancer development and spread.
How Deep Somatic Works
In clinical settings, scientists sequence tumor cells from biopsies as well as normal cells from the patient. Deep Somatic identifies differences, recognizing changes in tumor cells that are not inherited. These changes reveal factors fueling tumor growth.
The model transforms raw genome sequencing data from both cancerous and normal samples into images representing different data points, including sequencing data and their alignment along the chromosome. The convolutional neural network analyzes these images to distinguish between the standard reference genome, an individual’s natural inherited variants, and cancer-causing somatic variants while filtering out sequencing errors. The result is a list of cancer-associated mutations.
Training an Accurate AI Tool for Cancer Research
To train an accurate AI model, high-quality data is essential. Google and its partners at the University of California Santa Cruz Genomics Institute and the National Cancer Institute created a reference dataset called CASTLE. Tumor and normal cells were sequenced from four breast cancer samples and two lung cancer samples.
These samples were analyzed using three leading sequencing platforms to create an accurate reference dataset by integrating outputs and removing platform-specific errors. The data shows how the same type of cancer can have vastly different mutation signatures, information that can help predict a patient’s response to specific treatments.
Conclusion
Google’s Deep Somatic tool represents a significant advancement in accurately identifying cancer-associated genetic mutations. With this tool, researchers and doctors can better understand individual tumors, guide current treatments, and even develop new therapies. The goal is to enhance precision medicine and provide more effective treatments for patients, opening new horizons in the fight against cancer.