| CPC G16B 5/20 (2019.02) [G06N 5/022 (2013.01); G16B 40/00 (2019.02)] | 15 Claims |
|
1. A processor implemented method for preparing a knowledgebase of microbes and microbial function to reduce the risk of cancer in a person, the method comprising:
providing one or more microbial strains as a probiotic to a cancer host and a non-cancer host;
labelling, via one or more hardware processors, the one or more microbial strains into one of a good strain or a bad strain based on an impact on health of the cancer host and the non-cancer host;
performing, via a sequencer, a whole gene sequencing on the labelled good strain and the bad strains;
performing, via the one or more hardware processors, functional annotations on sequenced good strain and the bad strain to get a good and bad microbes' functional database;
filtering, via the one or more hardware processors, the good and bad microbes' functional database using a combination of a plurality of keywords in their names and the combination of the plurality of keywords are expected to pertain to microbial competition traits and have implications directly or indirectly in inhibiting or promoting cancer progression in collateral to competitive nature of microbes, wherein the filtering results in generation of a labelled knowledgebase of good and bad microbes with competition linked functional units;
performing, via the one or more hardware processors, genome clustering on the labelled knowledgebase of good and bad microbes using an unsupervised learning technique and validating if good and bad microbes are segregated on the basis of the competition linked functional units;
generating, via the one or more hardware processors, using a supervised machine learning technique:
a good microbe prediction model,
a bad microbe prediction model, and
functional units promoting competition against cancer;
learning, via the one or more hardware processors, a context whether a feature drives good microbe prediction or bad microbe prediction using Shapley additive explanations;
creating, via the one or more hardware processors, the knowledgebase of microbes and microbial function from the labelled knowledgebase of good and bad microbes containing only good prediction driving functional units;
receiving an unknown microbe;
identifying, via the one or more hardware processors, a nature of the unknown microbe using the context;
storing, via the one or more hardware processors, the knowledgebase of microbes and microbial function with the good microbes based on the identified nature;
creating a digital twin of a host-microbe-cancer crosstalk or interaction using the knowledgebase;
designing, via the one or more hardware processors, a cocktail of microbes, microbial molecules and microbe modulating molecules using the knowledgebase of microbes and microbial function to be given to the person at risk for use over an intermittent fasting regimen, immediately before and after a fasting period, wherein the cocktail is configured to avoid initiation, progression and side effects of cancer; and
administering the cocktail specific to a design meant for a specific type of cancer, wherein the administering of the cocktail is performed repeatedly for each intermittent fasting event that follows until one or more of the microbes have sustained in the host to yield a higher abundance in the fasting event as compared to a non-fasting event across multiple time points post cessation of the cocktail administration,
wherein the cocktail is administered in at least one of the form as a topical ointment to be applied on exposed surfaces vulnerable to cancer or affected by cancer, paste or liquid or gel or powder or a spray or a roller form of application involving delivery of a layer on top of an affected site, an injection or probiotic and prebiotic supplements,
wherein the processor implemented method is implemented for diagnosis or treatment of cancer without intervening an ongoing diagnosis or treatment.
|