Computerized system and method for integrated model-based drug discovery
Through a computerized framework based on decision intelligence, integrating artificial intelligence/machine learning models to analyze genomic data, the problem of insufficient accuracy in drug discovery and personalized treatment is solved, and higher drug selection and treatment effects are achieved.
Patent Information
- Application Number
- CN202380033587.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-03-10
- Filing Date
- 2023-03-09
- Publication Date
- 2024-11-22
AI Technical Summary
The prior art faces complex challenges posed by data digitization in drug discovery and personalized treatment, especially in drug efficacy prediction and patient stratification, with insufficient accuracy and robustness.
Using a computerized framework based on decision intelligence, the integrated algorithm is trained to analyze genomic data, improve the accuracy of drug selection and dose, and monitor drug pairs by integrating artificial intelligence/machine learning models, combining binary mutation data and hierarchical clustering data. Applicability of the patient's condition.
Significantly improved the accuracy and robustness of drug discovery and personalized treatment, with drug selection increasing responsiveness to specific patients by 7.7%, and expanded the functionality and applicability of drug selection and application through integrated models.
Smart Images

Figure CN119013727A_ABST
Abstract
Description
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] This application claims the benefit of priority to U.S. Provisional Patent Application No. 63 / 269,148, filed on March 10, 2022, which is incorporated by reference in its entirety. Technical Field
[0003] The present disclosure relates generally to drug efficacy prediction and patient stratification, and more particularly, to a computerized framework for drug discovery and suitability monitoring utilizing the application of decision intelligence (DI) based models. Background Art
[0004] The digitization of data in the pharmaceutical industry has grown dramatically over the past few years. However, this digitization involves, among other drawbacks, the challenges of acquiring, examining, and applying that knowledge to solve complex clinical problems. Summary of the invention
[0005] According to some embodiments, systems and methods for novel computerized frameworks for DI-based drug determination based on such digitization are disclosed. As discussed herein, the computerized framework can analyze genomic data and functions derived therefrom using an integrated configuration of artificial intelligence / machine learning (AI / ML) trained dynamically and recursively. According to some embodiments, integrated determination and application can increase the accuracy of training, validation, and external test sets associated with drug discovery and personalization. In addition, in some embodiments, an integrated computerized framework based on integration can be configured to analyze samples using an integrated algorithm trained using binary mutation data and hierarchical clustering data, which can enable determination of drug efficacy and patient stratification.
[0006] Thus, as is apparent from this disclosure, the use of an ensemble-based algorithmic approach provides improved accuracy compared to the use of a single algorithm (e.g., for the same data). For example, in predicting a patient's drug response, the ensemble approach can demonstrate a 7.7% increase in accuracy for drug selection and dosage identification for a particular patient. Thus, as discussed herein, more robust, accurate analysis and determinations are driven via an ensemble configuration.
[0007] As discussed herein, an ensemble model refers to a multi-layer combination of AI / ML (and / or deep learning) algorithms that combines the predictions of other algorithms as features for training the final algorithm and producing the final ensemble. The ensemble of predictors can be trained individually and their hyperparameters can be tuned to obtain optimal performance. Conceptually, the ensemble allows higher-level algorithms to learn from the performance of lower layers. In addition, the nature of the ensemble provides solutions to several common problems in machine learning. Therefore, when tuning a single algorithm, there may be a context in which one set of hyperparameters is optimal, and there may also be frequent anomalies that lead to inaccurate results. However, as discussed herein, in an ensemble, multiple algorithms (and / or multiple groups of algorithms) can be repeated many times with different tunings, thereby expanding the functionality and applicability of drug selection and application to patient conditions.
[0008] According to some embodiments, a method for a computerized framework based on DI to deterministically monitor and track the suitability of an identified drug for a patient's condition is disclosed. According to some embodiments, the present disclosure provides a non-transitory computer-readable storage medium for performing the above-mentioned technical steps of the functionality of the framework. The non-transitory computer-readable storage medium has computer-readable instructions tangibly stored thereon or tangibly encoded thereon, which, when executed by a device, causes at least one processor to perform a method for a computerized framework based on DI to deterministically monitor and track the suitability of an identified drug for a patient's condition.
[0009] According to one or more embodiments, a system is provided, the system comprising one or more processors and / or computing devices configured to provide functionality according to such embodiments. According to one or more embodiments, the functionality is embodied in steps of a method performed by at least one computing device. According to one or more embodiments, program code (or program logic) executed by a processor of a computing device to implement functionality according to one or more such embodiments is embodied in, by, and / or on a non-transitory computer-readable medium. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] Features and advantages of the present disclosure will be apparent from the following description of the embodiments as illustrated in the accompanying drawings, in which reference numerals refer to the same parts throughout the various views. The accompanying drawings are not necessarily drawn to scale, emphasis instead being placed upon illustrating the principles of the present disclosure:
[0011] Figure 1 is a block diagram of an example configuration within which the systems and methods disclosed herein may be implemented according to some embodiments of the present disclosure;
[0012] Figure 2is a block diagram illustrating components of an exemplary system according to some embodiments of the present disclosure; Figure 3 illustrates an exemplary workflow according to some embodiments of the present disclosure;
[0013] Figure 4 depicts non-limiting integrated model configurations according to some embodiments of the present disclosure;
[0014] Figure 5 Depicts an exemplary implementation of an architecture according to some embodiments of the present disclosure;
[0015] Figure 6 Describes an exemplary implementation of an architecture according to some embodiments of the present disclosure; and
[0016] Figure 7 is a block diagram of an exemplary computing device showing an example of a client or server device used in various embodiments of the present disclosure. DETAILED DESCRIPTION
[0017] The present disclosure will now be described more fully below with reference to the accompanying drawings, which form a part of the present disclosure and show certain example embodiments by way of non-limiting illustration. However, the subject matter may be embodied in various different forms, and therefore, the subject matter covered or claimed is intended to be interpreted as not being limited to any example embodiment set forth herein; the example embodiments are provided for illustration only. Likewise, it is intended to provide a reasonably broad scope for the subject matter claimed or covered. Among other aspects, for example, the subject matter may be embodied as a method, device, component, or system. Therefore, the embodiment may, for example, take the form of hardware, software, firmware, or any combination thereof (except the software itself). Therefore, the following detailed description is not intended to be understood in a restrictive sense.
[0018] Throughout the specification and claims, terms may have subtle meanings that are suggested or implied by the context beyond the explicitly stated meaning. Likewise, as used herein, the phrase "in one embodiment" does not necessarily refer to the same embodiment, and as used herein, the phrase "in another embodiment" does not necessarily refer to a different embodiment. For example, the claimed subject matter is intended to include, in whole or in part, a combination of the exemplary embodiments.
[0019] In general, terms may be understood at least in part based on usage in context. For example, as used herein, terms such as "and", "or" or "and / or" may include various meanings, which may depend at least in part on the context in which such terms are used. Typically, if used in an associative list, such as A, B, or C, "or" is intended to represent A, B, and C (used here in an inclusive sense), as well as A, B, or C (used here in an exclusive sense). In addition, as used herein, the term "one or more" may be used to describe any feature, structure, or characteristic in a singular sense, or may be used to describe a combination of features, structures, or characteristics in a plural sense, which depends at least in part on the context. Similarly, terms such as "a", "one", or "the / said" may be understood to convey singular use or to convey plural use, which depends at least in part on the context. In addition, the term "based on" may be understood to not necessarily be intended to convey a unique set of factors, and may alternatively allow for the presence of additional factors that are not necessarily explicitly described, again, which depends at least in part on the context.
[0020] The present disclosure is described below with reference to the block diagram and the operating instructions of method and equipment.It should be understood that the combination of each frame of the block diagram or operating instructions and the frame in the block diagram or the operating instructions can be realized by means of analog or digital hardware and computer program instructions.These computer program instructions can be provided to the processor of a general-purpose computer (to change its function, as described in detail herein), a special-purpose computer, an ASIC or other programmable data processing device, so that the instruction execution via the processor of a computer or other programmable data processing devices realizes the function / action specified in the block diagram or one or more operating frames.In some alternative embodiments, the function / action indicated in the frame may not occur in the order indicated in the operating instructions.For example, two frames shown continuously can actually be performed substantially simultaneously, or these frames can sometimes be performed in reverse order, depending on the function / action involved.
[0021] For the purposes of this disclosure, a non-transitory computer-readable medium (or one or more computer-readable storage media) stores computer data in a machine-readable form, which data may include computer program code (or computer-executable instructions) that can be executed by a computer. By way of example and not limitation, a computer-readable medium may include a computer-readable storage medium for tangible or fixed storage of data, or a communication medium for transient interpretation of a signal containing code. As used herein, a computer-readable storage medium refers to a physical or tangible storage device (as opposed to a signal), and includes (but is not limited to) volatile and non-volatile, removable and non-removable media implemented in any method or technology for tangible storage of information such as computer-readable instructions, data structures, program modules, or other data. Computer-readable storage media include, but are not limited to, RAM, ROM, EPROM, EEPROM, flash memory or other solid-state memory technology, optical storage devices, cloud storage devices, magnetic storage devices, or any other physical or material medium that can be used to tangibly store desired information or data or instructions and can be accessed by a computer or processor.
[0022] For purposes of this disclosure, the term "server" should be understood to refer to a service point that provides processing, database, and communication facilities. By way of example and not limitation, the term "server" may refer to a single physical processor with associated communications and data storage devices and database facilities, or it may refer to a complex of networked or clustered processors and associated network and storage devices, together with operating software and one or more database systems and application software that support the services provided by the server. A cloud server is an example.
[0023] For purposes of this disclosure, a "network" should be understood to refer to a network that can couple devices so that communications can be exchanged, such as between a server and a client device or other types of devices, including, for example, between wireless devices coupled via a wireless network. A network may also include, for example, mass storage devices such as a network attached storage device (NAS), a storage area network (SAN), a content delivery network (CDN), or other forms of computer or machine-readable media. A network may include the Internet, one or more local area networks (LANs), one or more wide area networks (WANs), a wired type connection, a wireless type connection, a cellular network, or any combination thereof. Likewise, subnetworks that may employ different architectures or may conform to or be compatible with different protocols may interoperate within a larger network.
[0024] For purposes of this disclosure, a "wireless network" should be understood as coupling a client device to a network. A wireless network may employ an independent ad hoc network, a mesh network, a wireless LAN (WLAN) network, a cellular network, and the like. A wireless network may also employ a variety of network access technologies, including Wi-Fi, Long Term Evolution (LTE), WLAN, a mesh of wireless routers, or second, third, fourth, or fifth generation (2G, 3G, 4G, or 5G) cellular technology, mobile edge computing (MEC), Bluetooth, 802.11b / g / n, and the like. The network access technology may enable wide area coverage for devices, such as, for example, client devices with varying degrees of mobility.
[0025] In short, a wireless network may include nearly any type of wireless communication mechanism by which signals may be transmitted between devices (such as client devices or computing devices), between networks, within a network, or the like.
[0026] A computing device is capable of sending or receiving signals, such as via a wired or wireless network, or is capable of processing signals or storing signals in memory as physical memory states, and can therefore operate as a server. Thus, as examples, a device capable of operating as a server may include a dedicated rack-mounted server, a desktop computer, a laptop computer, a set-top box, an integrated device combining various features (such as two or more of the foregoing devices), etc.
[0027] For purposes of this disclosure, a client (or user, entity, subscriber, or customer) device may include a computing device capable of sending or receiving signals, such as via a wired or wireless network. A client device may include, for example, a desktop computer or a portable device, such as a cellular phone, a smart phone, a display pager, a radio frequency (RF) device, an infrared (IR) device, a near field communication (NFC) device, a personal digital assistant (PDA), a handheld computer, a tablet computer, a tablet phone, a laptop computer, a set-top box, a wearable computer, a smart watch, an integrated or distributed device combining various features (such as those of the aforementioned devices), etc.
[0028] Client devices may vary in capabilities or features. The claimed subject matter is intended to cover a wide range of potential variations, such as, for example, a web-enabled client device or the previously mentioned device may include a high-resolution screen (e.g., HD or 4K), one or more physical or virtual keyboards, mass storage, one or more accelerometers, one or more gyroscopes, a global positioning system (GPS) or other location identification type capability, or a display with a high degree of functionality (such as a touch-sensitive color 2D or 3D display).
[0029] Certain embodiments and principles will be discussed in more detail with reference to the accompanying drawings. According to some embodiments, as discussed herein, the disclosed systems and methods provide computerized mechanisms for analyzing genomic data and functions derived therefrom. As discussed herein, an integrated computerized framework based on integration can be configured to analyze samples using an integrated algorithm trained with binary mutation data and hierarchical clustering data, which can enable determination of drug efficacy and patient stratification. In some embodiments, as discussed below, such drug efficacy and patient stratification can be based on a determined score corresponding to the responsiveness of the drug to the patient's condition.
[0030] refer to Figure 1 , depicting a system 100, which, according to some embodiments, may include a user equipment (UE) 102 (eg, a user device, as mentioned above and described below with respect to Figure 7 ), network 104, cloud system 106, database 108, and evaluation engine 200. It should be understood that although system 100 is depicted as including such components, this should not be construed as limiting, as one of ordinary skill in the art will readily appreciate that different numbers of UEs, cloud systems, databases, and / or networks may be utilized without departing from the scope of the present disclosure; however, for purposes of explanation, reference is made to the following. Figure 1 System 100 is discussed with reference to the example depicted in FIG.
[0031] According to some embodiments, UE 102 may be any type of device, such as, but not limited to, a mobile phone, a tablet, a laptop, a sensor, a wearable device, a wearable camera, a wearable garment, a patch, an Internet of Things (IoT) device, an autonomous machine, and any other type of modem device. In some embodiments, UE 102 may be a device associated with an individual (or a group of individuals) for whom drug responsiveness monitoring is provided.
[0032] In some embodiments, the UE 102 may be associated with a peripheral device (not shown), which may be connected to the UE 102 and may be any type of peripheral device, such as, but not limited to, a wearable device (e.g., a smart watch), a printer, a speaker, a sensor, etc. In some embodiments, the peripheral device may be a peripheral device that may be connected to the UE 102 via any type of known or yet to be known pairing mechanism (including, but not limited to, WiFi, Bluetooth, etc.). TM , Bluetooth Low Energy (BLE), NFC, etc.) connected to UE 102.
[0033] In some embodiments, network 104 may be any type of network, such as, but not limited to, a wireless network, a cellular network, the Internet, etc. (as discussed above). Network 104 facilitates connectivity of components of system 100, such as Figure 1 exemplified.
[0034] According to some embodiments, cloud system 106 can be any type of cloud operating platform and / or network-based system on which applications, operations, and / or other forms of network resources can be located. For example, system 106 can be a service and / or health provider and / or network provider by which services and / or applications can be accessed, supplied, or executed. For example, system 106 can represent a cloud-based architecture associated with a healthcare provider having associated network resources hosted on the Internet or a private network (e.g., network 104), which enables (via engine 200) patient monitoring and management discussed herein.
[0035] In some embodiments, the cloud system 106 may be a private cloud, where access is limited by isolating the network (such as preventing external access) or by using encryption to restrict access rights to only authorized users. Alternatively, the cloud system 106 may be a public cloud, where access is broadly available via the Internet. Public clouds may not be protected or may include limited healthcare features.
[0036] In some embodiments, the cloud system 106 may include a server and / or database of information accessible via the network 104. In some embodiments, the database 108 of the cloud system 106 may store data sets of data and metadata associated with users of the UE 102 and local and / or network information related to the UE 102, as well as services and applications provided by the cloud system 106 and / or the evaluation engine 200.
[0037] In some embodiments, for example, the cloud system 106 may provide a private / proprietary management platform whereby the engine 200 discussed below corresponds to novel functionality that the system 106 enables, hosts, and provides to the network 104 and other devices / platforms operating thereon.
[0038] Steering Figures 5 and 6 In some embodiments, the exemplary computer-based system / platform, exemplary computer-based device, and / or exemplary computer-based component of the present disclosure may be specifically configured to operate in a cloud computing / architecture 106, such as, but not limited to: Infrastructure as a Service (IaaS) 610, Platform as a Service (PaaS) 608, and / or Software as a Service (SaaS) 606 using a web browser, mobile application, thin client, terminal emulator, or other endpoint 604. Figures 5 and 6 Schematic diagram illustrating a non-limiting specific implementation of a cloud computing / architecture in which the exemplary computer-based system for managing control of custom and web-hosted APIs of the present disclosure may be specifically configured to operate.
[0039] Back to Figure 1According to some embodiments, database 108 may correspond to a data storage device for one platform (e.g., a network-hosted platform, such as cloud system 106, as discussed above) or multiple platforms. Database 108 may receive storage instructions / requests from, for example, engine 200 (and associated microservices), which may be in any type of known or to-be-known format, such as, for example, standard query language (SQL). According to some embodiments, database 108 may correspond to any type of known or to-be-known type of storage device, such as, but not limited to, a lookup table (LUT), a distributed ledger of a distributed network, etc.
[0040] As discussed above and further discussed in more detail below, the assessment engine 200 may include components for the disclosed functionality. According to some embodiments, the assessment engine 200 may be a dedicated machine or processor and may be hosted by a device on the network 104, within the cloud system 106, and / or on the UE 102. In some embodiments, the engine 200 may be hosted by a server and / or a group of servers associated with the cloud system 106.
[0041] According to some embodiments, as discussed in more detail below, the evaluation engine 200 can be configured to implement and / or control a plurality of services and / or microservices, wherein each of the plurality of services / microservices is configured to perform a plurality of workflows associated with performing the disclosed drug discovery and patient monitoring and management. Figure 3 Non-limiting embodiments of such workflows are provided.
[0042] According to some embodiments, as discussed above, the assessment engine 200 may be used as an application provided by the cloud system 106. In some embodiments, the engine 200 may be used as an application installed on a server, a network location, and / or other type of network resource associated with the system 106. In some embodiments, the assessment engine 200 may be used as an application operated via a conventional edge device (not shown) at a location associated with the system 100. In some embodiments, the engine 200 may be used as an application installed and / or executed on the UE 102. In some embodiments, such an application may be a web-based application accessed by the UE 102 from the cloud system 106 via the network 104. In some embodiments, the engine 200 may be configured and / or installed as an augmentation script, program, or application (e.g., a plug-in or extension) to another application or program provided by the cloud system 106 and / or executed on the UE 102.
[0043] like Figure 2As shown, according to some embodiments, the assessment engine 200 includes an identification module 202, an analysis module 204, a determination module 206, and an output module 208. It should be understood that the engines and modules discussed herein are non-exhaustive, as additional or fewer engines and / or modules (or sub-modules) may be applicable to embodiments of the systems and methods discussed. More details of the operation, configuration, and functionality of each of the engine 200 and its modules and their roles within embodiments of the present disclosure will be discussed below.
[0044] Go to Figure 3 , process 300 provides a non-limiting example implementation of the disclosed framework. According to some embodiments, process 300 provides a computerized mechanism for analyzing genomic data and functions derived therefrom. As discussed herein, the ensemble-based computerized framework can be configured to use an ensemble algorithm trained with binary mutation data and hierarchical clustering data (see discussion below). Figure 4 ) to analyze samples, which can enable determination of drug efficacy and patient stratification.
[0045] According to some embodiments, steps 302-304 and 314 of process 300 may be performed by identification module 202 of assessment engine 200. Steps 306 and 310 may be performed by analysis module 204; steps 308, 312, and 316 may be performed by determination module 206; and step 318 may be performed by output module 208.
[0046] According to some embodiments, process 300 begins at step 302, where engine 200 may identify the type of data and the corresponding data repository (e.g., a database that houses data of this particular type). For example, step 302 may involve engine 200 receiving a request from a user for identification of genomic data from public and / or private databases. In some embodiments, the data sets stored in such databases may include, but are not limited to, genomic data, mutation data, clinical outcome data (e.g., data representing, for example, treatment outcomes), treatment data, drug response data (e.g., half inhibitory concentration (IC50) data), molecular data, patient identifiers, demographics, biometrics, etc., or some combination thereof. Thus, such databases may store multiple distinct data sets of genomic data derived from multiple distinct diseased cells, each of which may include multiple pathway element data.
[0047] In step 304, the engine 200 may query the identified database to capture, retrieve, extract or otherwise identify the specifically requested type (and / or amount) of data. Thus, in step 304, the requested data may be retrieved or captured from the database.
[0048] In step 306, the engine 200 may analyze the captured data, wherein in step 308, a set of binary mutations may be determined based on the analysis of step 306. That is, in some embodiments, the engine 200 may receive the queried different data sets from the database, and identify the deterministic pathway elements (or features, both of which may be used interchangeably) in the different data sets that are associated with the status (e.g., sensitive or resistant) of the treatment parameters (e.g., treatment with a drug) of the diseased cells.
[0049] According to some embodiments, as discussed herein, pathway elements / signatures can be derived as continuous values based on the fraction of mutations that affect the same biological pathway from a binary value representing mutation status as 0 or 1. By way of example, if 4 mutations functionally contribute to the same biological function, and 2 of these mutations are mutated in a tumor sample, then for the pathway signature of that sample, a pathway signature with a fraction such as 2 / 4 or 0.5 can be created.
[0050] Thus, the engine 200 can generate information-rich features on a continuous value gradient, rather than many sparse data features, typically mostly 0 and a few 1. It is worth noting that mutations that occur at an average frequency may not be used in an AI / ML environment because, in a practical sample size (e.g., a sample size of 100), 5 mutations in a gene are too few for any currently known AI / ML algorithm to be trained by cross-validation and split out a test set (note that this may change when a more independent and less training-dependent AI / ML is discovered). As discussed herein, path feature engineering expands the number of features with sufficient variation by 10 to 50 times. The result is that there are more features available, and these have a higher amount of information. Therefore, relying on feature engineering (as provided herein) in addition to mutations (rather than just mutations) in an AI / ML environment enables prediction of drug response, which is not a feasible process without path information.
[0051] In some embodiments, when derived from a database rather than derived by clustering, path-based features can be created from paths organized as containing consistent effectors. For example, for some paths commonly used in biological analysis, it may be common to include activators and inhibitors of a single activity in the path gene list, which is not suitable for AI / ML feature calculation (for example, the derived path will be incoherent). For example, a score of 0.5 from this path can mean a mutation in 2 inhibitors or 2 activators, or 1 mutation each of an inhibitor and an activator. Their inconsistency will produce unclear features, which are not necessarily associated with a single functional impact. Therefore, a path with a consistent feature list (such as, for example, the human REACTOME database (reactome.org)) or a path from an experimental protein-protein interaction can be used as a source of a list for path feature calculation.
[0052] According to some embodiments, a score can be calculated from the raw binary mutations to represent the extent of binary mutations in a path, which can be normalized by the length of the path. This divides the sum of the binary mutations by the length of the available path to give a feature score in a sample. The available length is the length of the number of genes in the path after removing genes that were not available in the raw data.
[0053] By way of non-limiting example, in data from an assay where only 350 genes were measured, many pathways had genes that were not tested, and therefore these genes were removed from the calculation, resulting in an excessive indirect effect of the measurement technique on the feature weights. Thus, the disclosed methods can provide a mechanism for standardization across different techniques (e.g., any technique or technique where feature extraction and reduction can be utilized to focus on actionable elements for AI / ML processing). In some embodiments, for example, the pathway features are robust whether the genomic data consists of 300, 500, or more than 20,000 genes.
[0054] Thus, the query in step 304 may, for example, have conditions corresponding to, for example, elements of / used for pathway engineering (eg, pathway elements / features), states, and / or treatment parameters.
[0055] In some embodiments, in addition to the feature engineering discussed above, steps 306 to 308 may additionally (or alternatively, in some embodiments) involve analyzing a different data set associated with the diseased cells, and determining the path elements in the data set that are adjusted to produce the modified data set. Thus, in some embodiments, the engine 200 may utilize the modified data set to identify a state change in the treatment parameter of the diseased cells. In some embodiments, where desired and / or required, the engine 200 may pre-process the data set (e.g., feature selection, data transformation, metadata transformation, and / or split into a training data set and a validation data set) prior to the analysis of step 306.
[0056] In some embodiments, as described above, such analysis and determination (e.g., steps 306 to 308) may be performed by the engine 200 using any type of known or to-be-known AI / ML algorithm or technique, including but not limited to computer vision, classifiers, feature vector analysis, decision trees, boosting, support vector machines (SVM), neural networks (e.g., convolutional neural networks (CNN), recurrent neural networks (RNN), etc.), nearest neighbor algorithms, naive Bayes, bagging, random forests, logistic regression, etc.
[0057] Therefore, if Figure 4 As depicted, an integration 400 of such AI / ML algorithms may be utilized. As provided herein, in addition to other types of computational analysis discussed herein, integration 400 AI / ML combinations and / or iterations may enable more robust and accurate analysis of genomic data. In some embodiments, the integrated model may include, but is not limited to, an area under the curve formula, an R matrix (e.g., a matrix of applied algorithms, such as Figure 4 ), and p-value formulas (e.g., statistical and / or predictive measurement formulas). Thus, in some embodiments, integration achieves increased accuracy (or greater accuracy) compared to NIR (no information rate), as will be understood by those skilled in the art from the disclosure herein.
[0058] In some embodiments, and optionally, in any combination of the embodiments described above or below, the neural network technology can be, but is not limited to, one of a feedforward neural network, a radial basis function network, a recurrent neural network, a convolutional network (e.g., U-net), or other suitable networks. In some embodiments, and optionally, in any combination of the embodiments described above or below, the specific implementation of the neural network can be performed as follows:
[0059] a. Define the neural network architecture / model,
[0060] b. Transfer the input data to the neural network model,
[0061] c. Train the model incrementally,
[0062] d. determine the accuracy for a specific number of time steps,
[0063] e. Apply the trained model to process newly received input data,
[0064] f. Optionally, and in parallel, continue training the trained model at a predetermined periodicity.
[0065] In some embodiments, and optionally, in any combination of the embodiments described above or below, the trained neural network model may specify the neural network at least by a neural network topology, a series of activation functions, and connection weights. For example, the topology of the neural network may include the configuration of the nodes of the neural network and the connections between these nodes. In some embodiments, and optionally, in any combination of the embodiments described above or below, the trained neural network model may also be specified to include other parameters, including but not limited to bias values / functions and / or aggregation functions. For example, the activation function of a node may be a step function, a sine function, a continuous or piecewise linear function, a sigmoid function, a hyperbolic tangent function, or other types of mathematical functions representing the threshold at which a node is activated. In some embodiments, and optionally, in any combination of the embodiments described above or below, the aggregation function may be a mathematical function that combines (e.g., adds, multiplies, etc.) input signals to the node. In some embodiments, and optionally, in any combination of the embodiments described above or below, the output of the aggregation function may be used as an input to the activation function. In some embodiments, and optionally in a combination of any of the embodiments described above or below, the bias can be a constant value or function that can be used by the aggregation function and / or activation function to make a node more or less likely to be activated.
[0066] In some embodiments, the analysis and determination of steps 306 to 308 may include any nonlinear integration 400 of the AI / ML model, such as Figure 4 For example, such nonlinear combinations may include, but are not limited to, ensemble trees with bagging models, extreme gradient boosting, random forest models, SVM with Gaussian kernel models, elastic net logistic regression, at least power in the alpha band and power in the beta band, etc., and / or some combination thereof. In some embodiments, the features of the analyzed data may include, but are not limited to, at least BSR, standard deviation of FM, SVDE, and FD.
[0067] According to some embodiments, based on the type of data (from step 302) retrieved (in step 304), specific algorithms may be selected, which, as discussed above, may be tuned and / or configured for a specific type, format, and / or amount of data and computational analysis based on the data. In some embodiments, an AI / ML model may be applied to determine the type, identity, and / or number of AI / ML algorithms to utilize as part of integration 400. For example, for a particular BSR, bias, and SVDE, algorithms X and Y may be selected to iteratively operate on the subject data in a non-linear manner. Thus, different AI / ML algorithms may be specialized for identifying a portion of a solution; therefore, a combined approach may enable identification of a complete solution.
[0068] According to some embodiments, by way of non-limiting example, a customized algorithm integration (an example of which is provided in Figure 4 ) can be used with various types of data and model training objectives. For example, the customized AI / ML-based integration 400 can be applied to, but is not limited to: i) clinical patient mutation data, which may include information corresponding to cluster-derived module and pathway scores, and data corresponding to a classification training objective of patient response based on RECIST (response evaluation criteria in solid tumors); ii) clinical patient mutation data from the Cancer Genome Atlas (TCGA) glioblastoma group, which may include information corresponding to cluster-derived module and pathway scores, and data corresponding to a training objective of patient response based on increased survival above or below the median survival; iii) preclinical cell mutation data from cells of the National Cancer Institute (NCI) NCI60 panel, which does not include cluster-derived module and pathway scores and has a regression training objective of IC50 of cells treated with a drug (e.g., in -logIO IC50(M) format); iv) mutation data prepared from tumor samples, which can be scored in a binary manner (e.g., a sample is given 1 if a mutation occurs and 0 otherwise). In some embodiments, mutation data with synonymous mutations may not be replicated and therefore may be performed (eg, synonymous mutations for, eg, a patient, genome, and / or drug may be excluded and / or filtered out from the analyzed data).
[0069] Thus, in some embodiments, step 308 may therefore involve determining a score (or metric) for the data, thereby providing an indication of the value and / or amount of the binary mutation. As discussed above, such a score may be based on and / or involve the feature engineering process discussed above. As provided below, such a score may be used to group or cluster the data for subsequent drug responsiveness analysis.
[0070] In step 310, the engine 200 may analyze the determined data from step 308 (e.g., output from the integrated 400 analysis—e.g., determined binary mutations) and, in step 312, determine clusters of the data. According to some embodiments, such clustering may target (or group) one or more genes to identify features (e.g., pathway features) associated with responsiveness to one or more drugs and obtain hierarchical clustered data, as discussed below.
[0071] According to some embodiments, such analysis and clustering determination may be based on and / or involve the integrated determination and application discussed above with respect to steps 306 to 308. For example, based on the type of binary mutation, feature engineering, and / or criteria for clustering (e.g., the quantity and / or quality of the clustering data set), a specific AI / ML algorithm may be identified and executed in a non-linear and / or sequential manner.
[0072] By way of non-limiting example, the engine 200 can cluster the data by grouping (or "cutting" or filtering the data according to a particular criterion (e.g., the height or desired number of clusters). For example, if the criterion corresponds to k clusters, where k=40, then out of 400 binary mutations analyzed, 40 clusters may be generated.
[0073] According to some embodiments, in step 312, the engine 200 may perform clustering based on a set of features and / or biological pathways. In some embodiments, hierarchical clustering may be used to cluster the data to form clusters or groups, which enables the sparsity of the data to be reduced. In some embodiments, the data may be clustered based on the determined data scores from step 308, as discussed above. For example, mutations within a scoring range may be clustered together. In another non-limiting example, mutations may be grouped when their corresponding scores are equal to the sum of mutations associated with and / or indicating a particular biological pathway and / or such biological pathway.
[0074] According to some embodiments, in yet another non-limiting example, where there are initially 100 to 1100 columns of different features, the initial data may be selected or reduced using iterative feature reduction via an algorithm that is trained using hyperparameters tuned to optimize performance according to a set of constraints. In some embodiments, after evaluating this analysis, feature reduction may be performed when the evaluation fails to meet a performance threshold. In some embodiments, such feature reduction may occur in a stepwise manner until the total features (e.g., clusters) are less than or equal to the feature threshold (e.g., by gradually reducing 90% or 50% of the features to continue feature reduction until the number of available features is a predetermined percentage (e.g., 10%) of the total number of features (e.g., before reduction). Therefore, such feature reduction may prevent overfitting, thereby enabling the integration mechanisms and methods discussed herein to be effectively applied while ensuring improved accuracy when implemented.
[0075] When the data is clustered in step 312, the engine 200 may identify target drugs and target samples, as in step 314, whereby the clustered / reduced feature set (from step 312) may be fed to an integrated algorithm model in step 316. In some embodiments, the target samples correspond to and / or are represented by the clustered data. In some embodiments, the target samples and / or drugs may be identified from a database, as discussed above.
[0076] According to some embodiments, an ensemble algorithm model (e.g., which algorithms are part of an ensemble) may be configured and / or determined based on the type of clustering and / or the number of clusters, which may be determined in a manner similar to that discussed above. In some embodiments, a compiled ensemble from a previous step in process 300 may be identified and selected for use.
[0077] Thus, in step 316, the engine 200 may identify and / or determine an ensemble, whereby, when performing the ensemble, a determination regarding the responsiveness of the target drug may be performed. Thus, in step 316, the engine 200 may determine how the drug will work for the condition (e.g., represented via the target sample) (e.g., how effective the drug is). Such data may then be stored in a database, as in step 318. In some embodiments, such responsiveness data may be recursively fed back to the ensemble (or at least a portion of an algorithm used in the processing of process 300) to effect further training of the model implemented in the ensemble.
[0078] According to some embodiments, the input, i.e., a list of algorithms "R", such as, for example, "algo_name1", "algo_name2" (possibly with duplication (e.g., an ensemble may tune the same algorithm in different ways and execute the same algorithm in different ways, and both may contribute to the ensemble)) may be used as information in generating the ensemble discussed above, and / or may be associated with the stored information to provide an indicator of the analysis that led to the responsiveness determination. In some embodiments, with respect to training and / or retraining (e.g., via step 318), some customized steps may be performed to perform different data splits in the training data and 25% to 50% (optional) of the training data so the ensemble does not overfit (as discussed above). In some embodiments, different algorithms may be trained and saved in a "list" object in R, and then predictions using the model may be saved which may be added to data with patient samples and mutations and clusters. Thus, when the ensemble is stored and trained, the "final" algorithm selected for step 316 may be used to perform AI / ML training based on a new training set including the predictions stored in step 318.
[0079] Figure 7 is a schematic diagram of an exemplary client device, which shows an example implementation of a client device that can be used within the present disclosure. Client device 700 may include Figure 7 The components shown may be more or less than those shown. However, the components shown are sufficient to disclose an exemplary embodiment for implementing the present disclosure. The client device 700 may represent, for example, at least the above Figure 1 The UE 102 in question.
[0080] As shown, in some embodiments, the client device 700 includes a processing unit (CPU) 722 that communicates with a mass storage 730 via a bus 724. The client device 700 also includes a power supply 726, one or more network interfaces 750, an audio interface 752, a display 754, a keypad 756, an illuminator 758, an input / output interface 760, a tactile interface 762, an optional global positioning system (GPS) receiver 764, and a camera or other optical, thermal or electromagnetic sensor 766. As will be appreciated by those skilled in the art, the device 700 may include one camera / sensor 766 or multiple cameras / sensors 766. The power supply 726 provides power to the client device 700.
[0081] Client device 700 may optionally communicate with a base station (not shown), or directly with another computing device.In some embodiments, network interface 750 is sometimes referred to as a transceiver, a transceiver device, or a network interface card (NIC).
[0082] In some embodiments, the audio interface 752 is arranged to generate and receive audio signals, such as the sound of human speech. The display 754 can be a liquid crystal display (LCD), a gas plasma, a light emitting diode (LED), or any other type of display used with a computing device. The display 754 can also include a touch-sensitive screen arranged to receive input from an object (such as a stylus or a finger from a human hand).
[0083] Keypad 756 may include any input device arranged to receive input from a user. Illuminator 758 may provide status indications and / or provide light.
[0084] The client device 700 also includes an input / output interface 760 for communicating with the outside. In some embodiments, the input / output interface 760 can utilize one or more communication technologies, such as USB, infrared, Bluetooth, etc. TM Etc. The tactile interface 762 is arranged to provide tactile feedback to a user of the client device.
[0085] The optional GPS transceiver 764 can determine the physical coordinates of the client device 700 on the surface of the earth, which typically outputs the location as latitude and longitude values. The GPS transceiver 764 can also use other geolocation mechanisms, including but not limited to triangulation, assisted GPS (AGPS), E-OTD, CI, SAI, ETA, BSS, etc., to further determine the physical location of the client device 700 on the surface of the earth. However, in one embodiment, the client device can provide other information that can be used to determine the physical location of the device through other components, including, for example, a MAC address, an Internet Protocol (IP) address, etc.
[0086] The mass storage 730 includes RAM 732, ROM 734, and other storage components. The mass storage 730 illustrates another example of a computer storage medium for storing information, such as computer readable instructions, data structures, program modules, or other data. The mass storage 730 stores a basic input / output system ("BIOS") 740 for controlling the low-level operation of the client device 700. The mass storage also stores an operating system 741 for controlling the operation of the client device 700.
[0087] The memory 730 also includes one or more data repositories that can be utilized by the client device 700 to store, among other things, applications 742 and / or other information or data. For example, the data repositories can be used to store information describing various capabilities of the client device 700. This information can then be provided to another device based on any of a variety of events, including being sent as part of a header (e.g., an index file for an HLS stream) during communication, being sent upon request, etc. At least a portion of the capability information can also be stored on a disk drive or other storage medium (not shown) within the client device 700.
[0088] Application 742 may include computer executable instructions that, when executed by client device 700, transmit, receive and / or otherwise process audio, video, images, and enable remote communication with a server and / or another user of another client device. Application 742 may also include a client configured to send, receive and / or otherwise process games, goods / services and / or other forms of data, messages and content hosted and provided by a platform associated with engine 200 and its associated devices.
[0089] As used herein, the terms "computer engine" and "engine" identify at least one software component and / or a combination of at least one software component and at least one hardware component that is designed / programmed / configured to manage / control other software and / or hardware components (such as libraries, software development kits (SDKs), objects, etc.).
[0090] Examples of hardware elements may include processors, microprocessors, circuits, circuit elements (e.g., transistors, resistors, capacitors, inductors, etc.), integrated circuits, application specific integrated circuits (ASICs), programmable logic devices (PLDs), digital signal processors (DSPs), field programmable gate arrays (FPGAs), logic gates, registers, semiconductor devices, chips, microchips, chipsets, etc. In some embodiments, one or more processors may be implemented as complex instruction set computers (CISC) or reduced instruction set computers (RISC) processors; x86 instruction set compatible processors, multi-core or any other microprocessor or central processing unit (CPU). In various specific implementations, one or more processors may be dual-core processors, dual-core mobile processors, etc.
[0091] As used herein, computer-related systems, computer systems, and systems include any combination of hardware and software. Examples of software may include software components, programs, applications, operating system software, middleware, firmware, software modules, routines, subroutines, functions, methods, processes, software interfaces, application program interfaces (APIs), instruction sets, computer codes, computer code segments, words, values, symbols, or any combination thereof. Determining whether an implementation is implemented using hardware elements and / or software elements may vary according to any number of factors, such as desired computing rates, power levels, heat resistance, processing cycle budgets, input data rates, output data rates, memory resources, data bus speeds, and other design or performance constraints.
[0092] For purposes of this disclosure, a module is a software, hardware or firmware (or combination thereof) system, process or functionality, or component thereof that performs or facilitates the processes, features and / or functions described herein (with or without human-computer interaction or enhancement). A module may include submodules. The software components of a module may be stored on a computer-readable medium for execution by a processor. A module may be integrated with one or more servers, or loaded and executed by one or more servers. One or more modules may be grouped into an engine or application.
[0093] One or more aspects of at least one embodiment may be implemented by representative instructions stored in a computer-readable medium, which represents various logics within the processor, and when read by a machine, causes the machine to manufacture logic that performs the technology described herein. Such representations (referred to as "IP cores") may be stored on tangible machine-readable media and supplied to various customers or manufacturing facilities to be loaded into manufacturing machines that make logic or processors. It is worth noting that the various embodiments described herein can of course be implemented using any suitable hardware and / or computing software language (e.g., C++, Objective-C, Swift, Java, JavaScript, Python, Perl, QT, etc.).
[0094] For example, exemplary software specifically programmed according to one or more principles of the present disclosure can be downloaded from a network (e.g., a website) as a standalone product or as a plug-in package for installation in an existing software application. For example, exemplary software specifically programmed according to one or more principles of the present disclosure can also be available as a client-server software application or as a web-enabled software application. For example, exemplary software specifically programmed according to one or more principles of the present disclosure can also be embodied as a software package installed on a hardware device.
[0095] For the purposes of this disclosure, the term "user" or "patient" should be understood to refer to a user of one or more applications as described herein and / or a consumer of data provided by a data provider. By way of example and not limitation, the term "user" or "patient" may refer to a person who receives data provided by a data or service provider over the Internet in a browser session, or may refer to an automated software application that receives data and stores or processes the data. Those skilled in the art will recognize that the methods and systems of the present disclosure may be implemented in many ways and therefore should not be limited by the foregoing exemplary embodiments and examples. In other words, functional elements performed by a single or multiple components in the form of various combinations of hardware and software or firmware, as well as separate functions, may be distributed among software applications at the client level or the server level or both. In this regard, any number of the features of the different embodiments described herein may be combined into a single or multiple embodiments, and alternative embodiments with fewer or more than all the features described herein are possible.
[0096] Functionality may also be distributed in whole or in part among multiple components in ways now known or to be known. Thus, numerous software / hardware / firmware combinations are possible in implementing the functions, features, interfaces, and preferences described herein. Furthermore, the scope of the present disclosure encompasses conventionally known ways of implementing the described features and functions and interfaces, as well as those changes and modifications that may be made to the hardware or software or firmware components described herein, as will now and hereafter be understood by those skilled in the art.
[0097] In addition, the embodiments of the methods presented and described as flow charts in the present disclosure are provided by way of example in order to provide a more complete understanding of the technology. The disclosed methods are not limited to the operations and logical flows presented herein. Alternative embodiments are contemplated in which the order of the various operations is changed and in which sub-operations described as part of a larger operation are performed independently.
[0098] Although various embodiments have been described for the purposes of this disclosure, such embodiments should not be construed as limiting the teachings of this disclosure to those embodiments. Various changes and modifications may be made to the above elements and operations to obtain results that are still within the scope of the systems and processes described in this disclosure.
Claims
1. A method, include: The device identifies the genomic mutation data; Analyzing the genomic mutation data by the device via an integrated model; determining, by the device, a set of binary mutations based on the analysis; clustering, by the device, the set of binary mutations; analyzing, by the device, the clustered binary mutations based on information associated with the identified drug; determining, by the device, the responsiveness of each cluster to the identified drug based on an analysis of the clustered binary mutations; as well as Information indicative of the determined responsiveness is stored by the device in a database.
2. The method according to claim 1, further comprising: include: determining a set of algorithms based on said analysis of said genomic mutation data; as well as The integrated model is compiled based on the determined set of algorithms.
3. The method of claim 2, wherein the integrated model comprises a non-linear execution of the determined set of algorithms. The method of claim 2 , wherein the information stored in the database further indicates the set of algorithms. The method of claim 1 , wherein the ensemble is trained based on the determined responsiveness.
6. The method of claim 1, wherein the integration comprises a set of algorithms based on information including an area under the curve formula, an R matrix, and a p-value formula.
7. The method of claim 1, wherein the determined responsiveness comprises a score value, wherein the value indicates how the drug affects a particular cluster of binary mutations.
8. The method according to claim 1, further comprising: include: Identifying a type of each binary mutation in the binary mutations; The clustering of the binary mutations is performed based on the identified types, wherein each cluster corresponds to a specific type of binary mutation.
9. The method according to claim 1, further comprising: include: Feature reduction is iteratively performed on a set of binary mutation data as a step function until a set of features for the clusters satisfies a threshold, wherein the clusters are based on the feature reduction.
10. The method of claim 9, wherein the feature reduction is based on a set of constraints.
11. The method according to claim 1, further comprising: include: Identify the type of data and the corresponding database; querying the database; as well as The genomic mutation data is retrieved based on the query, wherein the identification of the genomic mutation data is based on the retrieval.
12. A device, include: A processor, the processor being configured to: Identify genomic mutation data; Analyzing the genomic mutation data via an integrated model; determining a set of binary mutations based on the analysis; clustering the set of binary mutations; analyzing the clustered binary mutations based on information associated with the identified drugs; determining the responsiveness of each cluster to the identified drug based on an analysis of the clustered binary mutations; as well as Information indicative of the determined responsiveness is stored in a database.
13. The apparatus of claim 12, wherein the processor is further configured to: determining a set of algorithms based on the analysis of the genomic mutation data; and The integrated model is compiled based on the determined set of algorithms, wherein the integrated model includes non-linear executions of the determined set of algorithms.
14. The apparatus of claim 12, wherein the processor is further configured to: Identifying a type of each binary mutation in the binary mutations; The clustering of the binary mutations is performed based on the identified types, wherein each cluster corresponds to a specific type of binary mutation.
15. The apparatus of claim 12, wherein the processor is further configured to: Feature reduction is iteratively performed on a set of binary mutation data as a step function until a set of features for the clustering satisfies a threshold, wherein the clustering is based on the feature reduction, wherein the feature reduction is based on a set of constraints.
16. The apparatus of claim 12, wherein the processor is further configured to: Identify the type of data and the corresponding database; querying the database; and The genomic mutation data is retrieved based on the query, wherein the identification of the genomic mutation data is based on the retrieval.
17. A non-transitory computer-readable storage medium tangibly encoded with computer-executable instructions, the computer-executable instructions when executed by a device, performing a method, the method include: Identifying genomic mutation data by the device; Analyzing the genomic mutation data by the device via an integrated model; determining, by the device, a set of binary mutations based on the analysis; clustering, by the device, the set of binary mutations; analyzing, by the device, the clustered binary mutations based on information associated with the identified drug; determining, by the device, the responsiveness of each cluster to the identified drug based on an analysis of the clustered binary mutations; as well as Information indicative of the determined responsiveness is stored by the device in a database.
18. The non-transitory computer readable storage medium of claim 17, further comprising: include: determining a set of algorithms based on said analysis of said genomic mutation data; as well as The integrated model is compiled based on the determined set of algorithms, wherein the integrated model includes non-linear executions of the determined set of algorithms.
19. The non-transitory computer readable storage medium of claim 17, further comprising: include: Identifying a type of each binary mutation in the binary mutations; The clustering of the binary mutations is performed based on the identified types, wherein each cluster corresponds to a specific type of binary mutation.
20. The non-transitory computer readable storage medium of claim 17, further comprising: include: Feature reduction is iteratively performed on a set of binary mutation data as a step function until a set of features for the clustering satisfies a threshold, wherein the clustering is based on the feature reduction, wherein the feature reduction is based on a set of constraints.
