| CPC G16B 40/00 (2019.02) [G01N 33/5008 (2013.01); G01N 33/5091 (2013.01); G16B 5/00 (2019.02); G16C 20/80 (2019.02); G16H 10/40 (2018.01); G16H 15/00 (2018.01); G16H 70/40 (2018.01); G16H 80/00 (2018.01)] | 15 Claims |
|
1. A computer-implemented method, the method comprising the steps of:
receiving, from a biological sample information input module associated with a network, biological sample information comprising genomic information from a plurality of subjects at a first time period, wherein the first time period is before a first drug treatment of cancer;
receiving, from the biological sample information input module associated with the network, biological sample information from the plurality of subjects at a second time period, wherein the second time period is during the first drug treatment with one or more drugs;
comparing, utilizing a biological sample information comparative analyzer module associated with the network, the biological sample information at the second time period with the biological sample information at the first time period;
generating, from a precedence graph generator module associated with the network, a precedence graph based on results of the comparison, wherein the precedence graph is generated by using frequencies of related pairs from a graph capturing a set of relationships from the comparing to identify those that meet a significance threshold from a given overall distribution of frequencies;
analyzing, utilizing the precedence graph generator module associated with the network, the precedence graph to determine if the biological sample information at the second time period comprises genomic alteration information of the biological sample for each of the plurality of subjects;
building, from a model builder module associated with the network, a model based on the analyzing the precedence graph utilizing a machine learning process;
implementing, based on a response actuator module associated with the network, a second drug treatment of cancer to one or more given subjects of the plurality of subjects based in part on the model;
comparing, utilizing the biological sample information comparative analyzer module associated with the network, biological sample information of a biological sample of another given subject different from the one or more given subjects of the plurality of subjects taken before a third drug treatment of cancer with the model based on analyzing the precedence graph to determine one or more probabilities of development of an alteration of the biological sample of the other given subject different from the one or more given subjects of the plurality of subjects; and
based on the one or more probabilities of development of an alteration of the biological sample, implementing, based on the response actuator module associated with the network, the third drug treatment to the other given subject that target the alteration;
wherein building, from the model builder module associated with the network, a model based on the analyzing the precedence graph utilizing a machine learning process, further comprises:
storing historical data of past performance for each biological sample information;
storing genomic alterations associated with a given drug treatment different from the first drug treatment, the second drug treatment and the third drug treatment of a particular drug or a drug combination that are potential targets for further drug treatment; and
creating a set of identified post-treatment mutational convergences associated with a resistance mechanism in response to the given drug treatment;
wherein implementing, based on the response actuator module associated with the network, the second drug treatment to the one or more given subjects of the plurality of subjects based in part on the model comprises identifying one or more other drugs different than the one or more drugs of the first drug treatment to improve response to the second drug treatment and decrease drug resistance in the second drug treatment; and
wherein the steps of the method are performed in accordance with a processor and a memory.
|