Analysis system and analysis method for tumor detection by metagenome sequencing data based on artificial intelligence

Abstract

The invention discloses an artificial intelligence-based analysis system for tumor detection by using metagenome sequencing data. The system comprises a data filtering module used for filtering low-quality sequences, a data comparison module used for comparing the sequences to a human reference genome, a chromosome fluctuation analysis module used for obtaining chromosome fluctuation variation conditions of a sample, and a tumor signal identification module based on artificial intelligence used for judging whether the sample contains tumor signals or not. The analysis system disclosed by the invention has the advantages of short analysis time and high accuracy, and can be used for detecting tumors by utilizing conventional mNGS sequencing data, so the function of detecting tumors while pathogens of one sample are detected is realized.

Description

technical field [0001] The invention belongs to the field of biological detection, and in particular relates to an analysis system and analysis method based on artificial intelligence for tumor detection using metagenomic sequencing data. Background technique [0002] Fever of unknown origin (FUO), commonly known as fever of unknown origin, generally refers to a group of diseases that have been fever for more than 3 weeks, and the body temperature has exceeded 38.3 ℃ for many times. With the development and promotion of metagenomic sequencing technology, in recent years, high-throughput sequencing technology is commonly used in clinical practice to sequence patient samples to find pathogens and investigate the cause of infection. In patients with fever under investigation, in addition to infection factors, tumors are often an important cause. In many cases, tumor screening is very difficult even with full-body scanning techniques such as CT, MRI, and PET. Many patients hav...

AI Technical Summary

Problems solved by technology

However, this judgment method has some limitations: 1. Different tumor samples have different fluctuation regions, lengths, and amplitudes, and it is difficult to cover all situations through a fixed threshold; 2. Some tumor samples do not have obvious fluctuating variation, but presents other more difficult-to-recognize chromosomal abnormalities, such as abnormalities at the ends of chromosomes ( figure 1 .b) and the chromosome fluctuation signal is slightly wavy, etc. ( figure 1 .c)

These chromosomal abnormalities are distinct from common fluctuations and thus cannot be detected by conventional fluctuation-variation correlation methods, prone to false negatives

3. Due to the difference in the source of the test sample, the test reagent used and the experimental operation process, the detected fluctuation signal pattern will also be different to a certain extent, such as the chromosome fluctuation caused by the change of sequence GC content, it is often difficult for people to distinguish it Distinguished from tumor-induced fluctuations ( figure 1 .d); 4. When a small-scale fluctuation is detected in the sample, it is difficult to distinguish whether it is a tumor fluctuation or a genetic variation

[0004] Based on the above problems, we realize that identifying tumor samples simply by detecting fluctuations and setting thresholds related to fluctuations cannot meet the actual clinical diagnosis needs.

Images