Computer-Implemented Method for Detecting an Analyte in an Immunoassay

Abstract

A computer-implemented system and method for detecting one or more analytes in an immunoassay is provided. The system and method of the disclosed embodiments can train a machine learning network using immunoassay reference data for the analytes and output a set of target experimental input parameters for detecting the analytes. The machine learning network can utilize the received data to generate target parameters and develop an immunoassay parameter set to detect and / or quantify the analyte. The computer-implemented system and method of the present disclosure can also generate a troubleshooting assistance output for the immunoassay.

Description

[Technical field]

[0001] (CROSS REFERENCE TO RELATED APPLICATIONS) This application claims the benefit of U.S. Provisional Patent Application No. 63 / 386,814, filed December 9, 2022, and U.S. Provisional Patent Application No. 63 / 295,160, filed December 30, 2021, each of which is incorporated by reference in its entirety herein.

[0002] FIELD OF THEINVENTION FIELD OF THE DISCLOSURE Embodiments of the present disclosure relate generally to computer-implemented methods for detecting and / or quantifying analytes, and more particularly to detecting and / or quantifying analytes by immunoassay methods. [Background technology]

[0003] Immunoassay experimental design techniques often utilize empirical testing methods to identify optimal parameters and solve problems. These empirical methods, such as trial and error testing, can be inefficient in terms of time and resources. For example, some immunoassay methods, such as immunoblotting or Western blotting, are not easily scalable due to pre-optimization limitations, and when multivariate optimization is introduced, optimization determinations quickly become more complex and time-consuming.

[0004] A key factor in optimizing immunoassay performance is determining the optimal reagent and experimental parameters to ensure successful detection in the immunoassay. Along with the analyte or protein concentration and antibody dilution, reagent selection is considered to ensure successful detection in an immunoassay experiment. The optimal values ​​of these variables are often determined experimentally.

[0005] Optimization techniques such as using best guess empirical values ​​and / or testing a range of values ​​for each individual experiment are very time consuming and expensive, but they are particularly common in Western blot experiments and workflows involving instruments such as imagers and blotters, including but not limited to iBright FL1500™, iBlot™, and chemiluminescent detection technologies.

[0006] Other factors such as optimization of immunoassay parameters based on analyte abundance, antibody binding variations, and selection of sensitivity detection substrates can introduce additional noise into the experimental data, resulting in misleading data, errors, and customer dissatisfaction. Additionally, noise can also be present in detecting analytes due to varying variables in protein preparations, protein forms, and experimental parameters when setting up and running the immunoassay. Summary of the Invention

[0007] The disclosed embodiments relate to computer-implemented systems and methods for detecting one or more analytes in an assay, such as an immunoassay. The disclosed embodiments of the systems and methods can use immunoassay reference data for the analytes to train a machine learning network and output a set of target experimental input parameters for detecting the analytes. The machine learning network can utilize the received data to generate target parameters and develop or generate an immunoassay parameter set for detecting and / or quantifying the analyte. The disclosed computer-implemented systems and methods can also generate troubleshooting assistance outputs for the immunoassay.

[0008] Embodiments relate to computer-implemented systems and methods for generating and outputting target parameters, such as, but not limited to, target experimental input parameters, for an immunoassay and / or for detecting one or more analytes in an immunoassay and / or for operating one or more immunoassay support systems and devices. In some embodiments, generating and outputting the target experimental input parameters is to improve or optimize the detection of an analyte by an immunoassay. Some embodiments utilize a semi-automated approach to extract data from standard immunoblot data, train a machine learning model using the validated immunoblot data, and apply a reference data set to generate and / or output and / or predict target experimental parameters for an immunoassay. Target parameters can include input variables such as sample concentration, protein concentration, detection reagents, and experimental input variables such as target loading concentration. Target loading concentration can aid in achieving a desired result and can be based on the type of experiment to be generated for a sample (e.g., a sample containing one or more analytes, a sample suspected of containing one or more analytes), a particular analyte, etc.

[0009] In some embodiments, systems and methods for detecting one or more analytes in an assay, such as an immunoassay, include training a machine learning network using immunoassay reference data comprising a set of reference input parameters and corresponding reference analyte detection data; receiving a set of experimental input parameters, the experimental input parameters including an identifier for the analyte; applying the machine learning network to the immunoassay reference data to identify target parameters based on the analyte; and determining a set of immunoassay parameters based on the target parameters, the set of immunoassay parameters including a loading concentration of at least one of the analyte and a detection reagent.

[0010] In various embodiments, the reference dataset includes experimental data from a large-scale antibody validation dataset. In some examples, the large-scale dataset can span more than 25,000 data points on a broad array of cell and tissue sets. Such data can be manually curated and can include sorted data demonstrating detection of proteins from cell or tissue lysates loaded across a range of loading concentrations, detection reagents, and antibody affinities.

[0011] Embodiments may further utilize machine learning and statistical learning models to generate and output target parameters and / or guide the user regarding target loading and reagent selection. Thus, results may provide insight into causal variables for optimizing immunoassay performance and generate recommendations regarding reagent selection, blocking agents, sample dilution ranges, antibody dilution ranges, and reagent dilution ranges, which may include, for example, protein loading, and detection reagents, among others.

[0012] Embodiments of the machine learning model uniquely integrate important reference data on cellular and tissue protein abundances, such as sample type abundances, with training datasets on experimental performance data across a broad set of immunoassays, and can further utilize experimental data with antibodies from multiple clonal and species-specific backbones. The technology can also utilize datasets with target binding affinities, loading values, multiple detection sensitivities, and cell lines to derive output values, such as, for example, binding changes at each protein abundance value.

[0013] Such techniques may be applied in clinical, research, academic, or diagnostic settings, among others. In various embodiments, the systems and methods may be executed on a computing device with a graphical user interface, such as a web tool, application, or other display. Such implementations may execute one or more of an experimental design tool and recommender for singleplex assays, troubleshooting features, feature extensions on experimental tools, user cloud accounts, troubleshooting modules, optimization modules, licensed executables for multiplex design, feature functionality with instrument control, and analysis software for multiplex assays. Implementations may further include an automated interface that can simulate detection of an immunoassay experiment and analyze theoretical experimental output.

[0014] Thus, the disclosed systems and methods can provide end users with information regarding target experimental parameters for target analyte detection and quantification. Exemplary analytes include, but are not limited to, haptens, hormones, nucleic acids, peptides, modified peptides, proteins, or modified forms of any of the above analytes. Tools for quantitative protein detection design can be automated and, according to other embodiments, can determine target loading values ​​of protein lysates for non-saturating signals. Various computational selection schemes for targeted experiment design can include an experiment design selector, and code for cell line selection.

[0015] Additional advantages and aspects of the disclosed technology include informing the end user through a software interface of the target protein loading for detection and quantification of proteins in an immunoblot assay, such as a Western blot assay. The output can inform the range of detection to provide recommendations regarding antibody concentration and detection reagent selection for optimal detection of proteins, such as detection within the target range. In embodiments, the target range can be defined based on one or more factors, including but not limited to user selection, visibility, discrimination, or range to achieve a particular result, such as analyte, or other defined criteria. Thus, these tools can inform the user of optimal variables for selection and design of multiplexed protein detection in an immunoassay, aiding in experimental design, optimization, and troubleshooting.

[0016] Embodiments of the disclosed technology also include manual and automated image analysis tools for extracting protein detection values ​​from immunoassay and immunoblot data. Such embodiments can match protein detection to reference values ​​of protein abundance or transcript abundance. The system can apply methods to clean and fit data to eliminate outlier data on protein instability and classify data on variables such as antibody clonality across different cell lines, detection sensitivity, loading concentration, and protein abundance. Datasets can be modeled, trained, and tested using one or more statistical models, machine learning models, or a combination thereof.

[0017] In certain examples, embodiments of the disclosed technology can extract protein detection information from standard immunoblot experiments such as Western blots. Embodiments can model protein abundance, clonality, detection, and select multiplexed protein detection and target loading detection values.

[0018] A computer-implemented method for analyte detection in an immunoassay can include receiving immunoassay reference data including a set of reference input parameters and corresponding reference analyte detection data, receiving a set of experimental input parameters, the experimental input parameters including an analyte, applying a machine learning network to the immunoassay reference data to identify target parameters based on the analyte, and determining a set of immunoassay parameters based on the target parameters, the output immunoassay parameter set including a loading concentration of at least one of the analyte and the detection reagent. In some embodiments, the loading concentration of at least one of the analytes includes a loading concentration of a sample including the analyte. In some embodiments, the loading concentration of at least one of the analytes includes a loading concentration of a sample including the analyte, the sample may be in a sample buffer. In some embodiments, the loading concentration of at least one of the analytes includes a loading concentration of the analyte included in a buffer or solution or mixture.

[0019] In various embodiments, the immunoassay reference data includes at least one of Western blot data, multiple Western blot captures, quantitative data, optionally quantitative data representing relative abundance of proteins, categorical data, protein detection data, and immunocytometry data.

[0020] The training input parameters can include one or more of user input, analyte type, protein type, clonality, cell line, detection reagent, antibody concentration, substrate, substrate sensitivity, detection sensitivity, lysate concentration, cell line, set of proteins, antibody binding change, blocking data, cell lysate preparation data, protein instability, protein stability, gel parameters (e.g., gel composition, gel porosity, etc.), analyte amount range, protein mass range, cell line, detection data, lysate type, lysate loading concentration, protein, protein isoform, fragment of protein or post-translationally modified protein, antibody binding site, antibody clonality, antibody concentration / dilution, binding affinity, antibody isoform specificity, scaffold type, protein stability, protein instability, detection label, enzyme, detection multiplicity, or detection sensitivity. In an embodiment, the training input parameters can include user input. In a singleplex experiment, for example, the training input can include protein type, cell line, lysate concentration, and antibody dilution. In multiplex experiments and troubleshooting experiments, the training input can include user input indicating variables such as a set of proteins. In a modeling operation, the training input can include a set of ranges of variables. The variables may relate to one or more of the following: experiment-related parameters, desired inputs or outputs, and other information related to the immunoassay experiment or apparatus.

[0021] An immunoassay parameter set can represent the output or prediction or generation of one or more variables, settings, and experimental inputs for performing an immunoassay experiment. The immunoassay parameter set can include, for example, one or more of cell line, analyte loading range, target protein, detection data, lysate type, lysate loading concentration, antibody clonality, antibody type, antibody binding site, antibody dilution range, sample dilution range, reagent dilution range, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection reagent, detection multiplicity, detection sensitivity, target range of loading concentration of at least one of the analytes (in some embodiments the analyte is contained in the sample and / or contained in a buffer, sample buffer, solution, and / or mixture), clonality, antibody type, detection reagent, immunoassay performance prediction, recommended protein source, optimal cell line, cell source, antibody clonality, detection technique, clonality recommendation, antibody recommendation, analyte recommendation, analyte source recommendation, gel type, generated or output or recommended detection reagent, antibody type, antibody loading concentration / dilution, protein lysate concentration, target cell line, target protein source, analyte amount, transfer conditions, validation flags, and predicted analyte localization. In embodiments, an analyte can be located in multiple locations, for example, an analyte can be secreted from a cell, can be present in an extracellular fluid, can be recombinantly expressed, can be targeted to a non-native location, can be present intracellularly in one or more cellular compartments, can be biochemically fractionated, and / or can be resolved with respect to biophysical / biochemical properties (such as Western blot or isoelectric focusing).

[0022] A system for detecting one or more analytes in an immunoassay comprises at least one computing device comprising a processor and at least one memory storing instructions that, when executed by the processor, cause the computing device to receive immunoassay reference data comprising a set of input parameters and corresponding analyte detection data; receive a set of experimental input parameters, the experimental input parameters comprising an analyte identifier and clonality; apply a machine learning network to the immunoassay reference data to identify target parameters based on the analyte; determine an immunoassay parameter set based on the target parameters; and determine an immunoassay parameter set based on the target parameters, the immunoassay parameter set comprising a loading concentration of at least one of an analyte and a detection reagent contained in a sample, a buffer, a mixture, and / or a solution.

[0023] In some embodiments, in the systems of the present disclosure, the at least one memory stores instructions that, when executed by the processor, further cause the computing device to extract relevant immunoassay reference data based on the experimental input parameters, determine a relationship between the experimental input parameters and the corresponding reference analyte detection data, and determine or generate target parameters for detecting the analyte.

[0024] In some embodiments, in the systems of the present disclosure, the at least one memory stores instructions that, when executed by the processor, cause the computing device to further classify the associated immunoassay reference data into variables, apply a statistical model to determine a relationship between two or more variables, and train a machine learning network using the relationship between the two or more variables.

[0025] In some embodiments, in a system of the present disclosure, the instructions further cause the computing device to extract immunoassay reference data from the image representative of the experimental immunoassay data.

[0026] The system of the present disclosure, in some embodiments, further includes a user interface, the user interface comprising at least one of an instrument console, a web tool, a graphical user interface, and a display on a computing device.

[0027] In some embodiments, the present disclosure comprises a non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause a device to perform one of the methods described herein.

[0028] In some embodiments, a computer-implemented method of the present disclosure is for operating an immunoassay instrument support apparatus, the method including: receiving immunoassay reference data including a set of reference input parameters and corresponding reference analyte detection data; receiving a set of experimental input parameters, the experimental input parameters including at least one variable associated with an immunoassay experiment; receiving information indicative of a problem related to at least one of the immunoassay reference data, the experimental input parameters, and a result of the immunoassay experiment; applying a machine learning network to the immunoassay reference data to identify target parameters based on the problem; and determining a solved immunoassay parameter set based on the target parameters.

[0029] Exemplary problems include, but are not limited to, at least one of an analyte detection problem or an immunoassay reference data extraction problem. In some embodiments, the problem relates to extraction of immunoassay reference data from an image showing experimental immunoassay data.

[0030] In some embodiments, a computer-implemented method of the present disclosure is for operating an immunoassay instrument supporting apparatus, comprising: receiving immunoassay reference data including a set of reference input parameters and corresponding reference analyte detection data; receiving a set of experimental input parameters; receiving at least one target variable for an immunoassay experiment; applying a machine learning network to the immunoassay reference data to identify a target parameter based on the target variable; and determining a set of immunoassay parameters for obtaining the target variable based on the target parameters, where the target variable is an analyte. In embodiments, the set of experimental input parameters represents a method of experimentation. Non-limiting methods of experimentation include Western blot, immunoblot, transfer, and protein gel-based methods.

[0031] In some embodiments, a computer-implemented method of the present disclosure for operating an immunoassay instrument support apparatus for immunoblotting includes receiving immunoblot reference data comprising a set of reference input parameters and corresponding reference analyte detection data; receiving a set of experimental input parameters, the experimental input parameters including an identifier for the analyte; applying a machine learning network to the immunoblot reference data to identify target parameters based on the analyte; and determining a set of immunoassay parameters based on the target parameters, the set of immunoassay parameters including a loading concentration of at least one of an analyte and a detection reagent in a sample, a buffer, a mixture, and / or a solution.

[0032] In some embodiments, a computer-implemented method of the present disclosure for operating an immunoassay instrument support apparatus for immunoblotting includes receiving immunoblot reference data comprising a set of reference input parameters and corresponding reference analyte detection data; receiving a set of experimental input parameters, the experimental input parameters including an identifier for the analyte; applying a machine learning network to the immunoblot reference data to identify target parameters based on the analyte; and determining a set of immunoassay parameters based on the target parameters, the set of immunoassay parameters including a loading concentration of at least one of an analyte and a detection reagent in a sample, a buffer, a mixture, and / or a solution.

[0033] In some embodiments of the present disclosure, a method for operating an immunoassay instrument support apparatus includes receiving immunoassay reference data including a set of reference input parameters and corresponding reference analyte detection data, receiving a set of experimental input parameters, the experimental input parameters including an analyte identifier and single cell information, applying a machine learning network to the immunoassay reference data to identify target parameters based on the analyte identifier, and determining a set of immunoassay parameters for detecting multiple proteins in the single cell based on the target parameters, the set of immunoassay parameters including a loading concentration of at least one of an analyte and a detection reagent contained in the sample. Non-limiting examples of single cells include one or more of a cell line or cell lineage identified by microscopic morphology and sorted / tracked using one or more biomarker tags, cells expressing a fluorescent reporter, or cells cultured as isolated primary or maintained cell lines. Non-limiting examples of single cell information may include one or more of transcript abundance, analyte (e.g., protein, nucleic acid, etc.) localization information, or proteomic data on analyte / protein abundance (e.g., mass spectrometry studies on analyte / protein abundance), protein post-translational modifications (e.g., glycan modifications), phosphorylation, ubiquitination, sumoylation, lipid anchors, and the like.

[0034] Some embodiments describe a computer-implemented method for operating an immunoassay instrument supporting apparatus for a flow-based immunoassay, the method including: receiving immunoassay reference data including a set of reference input parameters and corresponding reference analyte detection data, the immunoassay reference data including at least one set of flow-based immunoassay data; receiving a set of experimental input parameters, the experimental input parameters including an analyte identifier and single-cell information; applying a machine learning network to the immunoassay reference data including the flow-based immunoassay data to identify target parameters based on the analyte; and determining an immunoassay parameter set for detecting a plurality of proteins in a single cell based on the target parameters, the immunoassay parameter set including a loading concentration of at least one of an analyte and a detection reagent in the sample.

[0035] In some embodiments, the single cell information can be data about surface markers, secreted analytes, or intracellular markers. Such single cell information can be obtained from flow cytometry, image analysis, or protein localization studies.

[0036] Exemplary non-limiting flow assays include assays in which a set of analytes / proteins with different abundances are profiled, with the matched dye intensity being inversely proportional to the abundance value of the analyte / protein. In such an example, one embodiment involves identifying cells whose cell types express a set of analytes (proteins or protein modification targeting antibodies) and suitable for simultaneous interrogation of a user-selected set of analytes in an experiment. Another embodiment is to select a set of antibodies and dyes that provide optimal spectral compensation based on dye intensity and expected analyte abundance.

[0037] Thus, embodiments of the disclosed technology include systems and methods for applying machine learning to improve detection in immunoassays and / or for inputting input parameters and receiving an output of a set of target immunoassay parameters, e.g., via a user interface, applying various techniques to a computer, web tool, or other hardware device, and / or troubleshooting problems. Such techniques may be applied to Western blotting, bead-based assays, multi-analyte profiling, multi-analyte detection, multi-protein profiling, multi-protein detection, analyte imaging, protein imaging, flow cytometry-based detection, and fluorescent and chromogenic detection. [Brief description of the drawings]

[0038] The Summary and the following Detailed Description will be better understood when read in conjunction with the accompanying drawings. For purposes of illustrating the disclosed subject matter, exemplary embodiments of the disclosed subject matter are shown in the drawings, but the disclosed subject matter is not limited to the particular methods, compositions, and devices disclosed. Additionally, the drawings are not necessarily drawn to scale. In the drawings, [Figure 1] 1 illustrates an exemplary flowchart for providing recommendations in accordance with an embodiment of the disclosed technology. [Diagram 2] 1 shows an exemplary flow chart for predicting a target parameter for analyte detection, according to an embodiment of the disclosed technology. [Diagram 3] 1 shows a model relationship between protein abundance, clonality, and detection according to an embodiment of the disclosed technology. [Figure 4] 1 shows an immunoassay dose response curve according to an embodiment of the disclosed technology. [Diagram 5] 1 illustrates linear and logistic dose-response models according to embodiments of the disclosed technology. [Figure 6] 1 illustrates an implementation of a machine learning model in accordance with an embodiment of the disclosed technology. [Figure 7]1 shows a multiplex protein detection flow chart according to an embodiment of the disclosed technology. [Figure 8] 1 illustrates an exemplary user interface output in accordance with an embodiment of the disclosed technology. [Figure 9] 1 illustrates an exemplary user interface output in accordance with an embodiment of the disclosed technology. [Figure 10] 1 shows an immunoassay experiment according to an embodiment of the disclosed technology. [Figure 11] 1 shows an immunoassay experiment according to an embodiment of the disclosed technology. [Figure 12] 1 shows an immunoassay experiment according to an embodiment of the disclosed technology. [Figure 13] 1 shows an immunoassay experiment according to an embodiment of the disclosed technology. [Figure 14] 1 shows an immunoassay experiment according to an embodiment of the disclosed technology. [Figure 15] 1 shows target abundance based predictions according to embodiments of the disclosed technology. [Figure 16] 1 illustrates a computing system in accordance with an embodiment of the disclosed technology. [Figure 17] 1 illustrates a method for identifying an analyte-based target parameter of an experiment using an embodiment of the disclosed technology. [Figure 18] 1 shows a method for determining cell line recommendations for use in immunoassay experiments using embodiments of the disclosed technology. [Figure 19] 1 shows a method for ranking cell line recommendations for use in immunoassay experiments using embodiments of the disclosed technology. [Figure 20] 1 shows a method for determining cell line recommendations for use in immunoassay experiments using embodiments of the disclosed technology. [Figure 21] 1 illustrates an example of a user interface that receives analytes of interest and displays cell line recommendations for immunoassay experiments, according to embodiments of the disclosed technology. [Figure 22]1 shows an example of a user interface displaying a ranked order of cell line recommendations for immunoassay experiments, according to embodiments of the disclosed technology. [Diagram 23] 1 shows an example of a user interface that recommends experiment recommendations regarding specific features of an immunoassay experiment, according to an embodiment of the disclosed technology. [Figure 24] 1 shows an example of a user interface that displays recommendations for specific features of an immunoassay experiment, according to an embodiment of the disclosed technology. [Diagram 25] A method for determining analyte migration in an immunoassay using an embodiment of the disclosed technology. [Figure 26] 1 shows an exemplary diagram of an immunoassay experiment, according to an embodiment of the disclosed technology. [Figure 27] 1 illustrates a method for a system that can be used to determine band migration in immunoassay experiments using embodiments of the disclosed technology. [Figure 28] 1 shows an example of a user interface for a user to load immunoblot data for analysis of immunoblots and analyte migration according to embodiments of the disclosed technology. [Figure 29A] 1 shows an example of a user interface for receiving experimental data for an immunoassay experiment using embodiments of the disclosed technology. [Figure 29B] 1 shows an example of a user interface for receiving experimental data for an immunoassay experiment using embodiments of the disclosed technology. [Figure 30A] 1 shows an example of a user interface displaying analytical results of an immunoassay experiment using an embodiment of the disclosed technology. [Figure 30B] 1 shows an example of a user interface displaying analytical results of an immunoassay experiment using an embodiment of the disclosed technology. [Diagram 31] 1 shows an example of an immunoblot with multiple bands of interest that have been reported using embodiments of the disclosed technology.

[0039] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention are now described with reference to the drawings, in which the drawing in which an element first appears is typically indicated by the leftmost digit(s) in the corresponding reference number. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0040] The present disclosure may be understood more readily by reference to the following detailed description taken in conjunction with the accompanying drawings and examples which form a part of this disclosure: It is understood that the present disclosure is not limited to the specific devices, methods, applications, conditions or parameters described and / or illustrated herein, and that the terminology used herein is for the purpose of describing particular embodiments by way of example only, and is not intended to limit the claimed subject matter.

[0041] In the following detailed description, references to "one aspect," "aspect," "exemplary aspect," etc., indicate that the aspect being described may include a particular feature, structure, or characteristic, but not all aspects necessarily include the particular feature, structure, or characteristic. Moreover, such phrases do not necessarily refer to the same aspect. Furthermore, when a particular feature, structure, or characteristic is described in connection with an aspect, it is submitted that it is within the knowledge of one of ordinary skill in the art to achieve such feature, structure, or characteristic in connection with other aspects, whether or not explicitly described. The terms "aspect" and "embodiment" may be used interchangeably.

[0042] Also, as used in the specification, including the appended claims, the singular forms "a," "an," and "the" include the plural, and a reference to a particular numerical value includes at least that particular numerical value, unless the context clearly dictates otherwise. As used herein, the term "plural" means two or more. When a range of values ​​is expressed, another embodiment includes from the one particular numerical value and / or to the other particular numerical value. Similarly, when numerical values ​​are expressed as approximations, by use of the antecedent "about," it will be understood that the particular numerical value forms another embodiment. All ranges are inclusive and combinable. It is understood that the terminology used herein is for the purpose of describing particular aspects only and is not intended to be limiting.

[0043] It should be understood that, for clarity, certain features of the disclosed subject matter that are described herein in separate embodiments may also be provided in combination in a single embodiment. Conversely, various features of the disclosed subject matter that are, for brevity, described in the context of a single embodiment, may also be provided separately or in any subcombination. Furthermore, references to values ​​stated in ranges include all values ​​within that range. All documents cited herein are incorporated by reference in their entirety for all purposes.

[0044] Immunoassays can be used to detect the presence and / or expression level of a particular analyte (e.g., peptide, polymer, protein, etc.) in a sample. Immunoassays are test experiments that use tagging molecules to detect and quantify substances, specifically analytes, in a test sample. In some aspects, the tagging molecules may be antibodies, oligonucleotides, or analyte stains (e.g., Coomassie, silver stain, etc.). Often, immunoassays use electrophoresis (e.g., gel electrophoresis) to detect and quantify specific analytes. The output of an immunoassay may be an image called an immunoblot. It is expected that analytes migrate through a gel (or other membrane, depending on the type of electrophoresis used) based on a function of mass (molecular weight), such that analytes of different masses can be identified as distinct bands in the immunoblot. Such techniques can be used, for example, to identify and quantify proteins during protein preparation.

[0045] Embodiments of the present disclosure describe systems and methods for detecting one or more analytes in an assay, such as an immunoassay, including training a machine learning network using immunoassay reference data including a set of reference input parameters and corresponding reference analyte detection data, receiving a set of experimental input parameters, the experimental input parameters including an identifier for the analyte, applying the machine learning network to the immunoassay reference data to identify target parameters based on the analyte, and determining a set of immunoassay parameters based on the target parameters, the set of immunoassay parameters including a loading concentration of at least one of the analyte and a detection reagent. Methods of embodiments of the present disclosure include computer-implemented methods.

[0046] The embodiments relate to systems and methods for generating and predicting target parameters, such as, but not limited to, target experimental input parameters, for immunoassays and / or for detecting one or more analytes in an immunoassay and / or for operating one or more immunoassay support systems and devices. In some embodiments, generating and predicting target experimental input parameters is optimizing assay conditions for an immunoassay. Some embodiments utilize a semi-automated approach to extract data from standard immunoblot data, train a machine learning model using the validated immunoblot data, and apply a reference data set to predict target experimental parameters for an immunoassay. Target parameters can include input variables such as sample concentration, protein concentration, detection reagents, and experimental input variables such as target loading concentration. Target loading concentration can aid in achieving a desired result and can be based on the type of experiment to be generated for a sample (e.g., a sample containing one or more analytes, a sample suspected of containing one or more analytes), a particular analyte, and the like.

[0047] Non-limiting examples of samples having analytes detected or quantified by the methods and systems of the present invention include biological samples, water samples, environmental samples, air samples, forensic samples, agricultural samples, pharmaceutical samples, food samples. Analytes can be natural or synthetic. Biological samples can be samples obtained from eukaryotic or prokaryotic sources. Non-limiting examples of eukaryotic sources include mammals, such as humans, cows, pigs, chickens, turkeys, livestock animals, fish, crabs, crustaceans, rabbits, game animals, and / or murine animals, such as rats or mice. Biological samples can include non-limiting examples of biological fluids, cells and / or tissues, including blood, plasma, cerebrospinal fluid, lymph, bone marrow, nasal fluid / swabs, throat swabs or samples, saliva, urine, feces, cell samples (including cells, single cells, cell lysates, cell components, and / or material derived from cells), or tissues.

[0048] Embodiments of the disclosed technology can provide output of target parameters (such as target experimental input parameters) to improve analyte detection in assays such as immunoassays. In some embodiments, the technology to output parameters can utilize a semi-automated approach. Embodiments further include methods to extract data from standard immunoblot data, train a machine learning network or model using the validated immunoblot data, and apply a reference data set to output experimental parameters for a target or improved immunoblot assay.

[0049] Embodiments in accordance with the disclosed technology include systems and methods for applying machine learning to improve detection in immunoassays, providing or outputting target parameters, and providing experimental design parameter output or recommendations based on desired input parameters, methods for profiling multiple analytes such as, but not limited to, proteins, methods for detecting multiple analytes such as, but not limited to, proteins in single cell experiments, and systems and methods for various applications to Western blot experiments, flow-based detection, fluorescent dye detection, and tag counting-based detection methods.

[0050] Thus, embodiments can uniquely integrate important reference data on protein abundance across cells and tissues with training datasets on experimental performance data across a broad set of immunoblotting assays, experimental data with antibodies from multiple clonal / species-specific scaffolds with target binding affinity can be used to train models and derive binding changes at each protein abundance value, and data from multiple detection sensitivities, loading values, and cell lines can be used to model and predict output values. In examples, such important reference data on protein abundance per cell and tissue can refer to cell and protein amounts, cell or tissue types used in previous experiments, as discussed herein.

[0051] The embodiment may include a validation selector and code for cell line selection and automation. The embodiment additionally has the ability to apply large training sets and analyze multiplexed protein detection using advanced machine learning and AI-based decision models with an easy to use interface design.

[0052] As described herein, multiplex assay refers to an immunoassay that can measure multiple analytes. Thus, multiplex protein detection refers to the detection of multiple proteins. In some aspects, multiplex assays can utilize pull-down techniques, magnetic beads, for example, to bind antibodies, cells, or lysate preparations. In some aspects, multiplex protein detection can utilize antibodies analyzed using IP-MS, secondary antibody enzymes, dye conjugates, primary conjugated dyes, or polymer tags. It is understood that a variety of multiplex assays, proteins, sources, analytes, and samples known in the art can be used according to the embodiments described herein.

[0053] Thus, the disclosed technology is uniquely able to at least predict immunoassay performance, design immunoassays that are useful in using these technologies in clinical and diagnostic settings, optimize protein quantification and detection, and optimize loading and detection sensitivity for protein quantification.

[0054] The disclosed technology also supports experimental planning and design techniques for advancing growth in multiplexed protein detection and precision cellular analysis, for example, through the ability to optimize and troubleshoot immunoblot assays, determine target loading of protein lysates for detection in immunoblot experiments, and design multiplexed western detection in cell or tissue lysates.

[0055] FIG. 1 shows an exemplary flow chart for embodiments and models for providing immunoassay design recommendations as described herein. At 110, an embodiment can receive immunoassay reference data including a set of input parameters that can correspond to reference analyte detection data. The immunoassay reference data can be derived from previous experiments and can include results from the set of input parameters. At 120, an embodiment can receive a set of experimental input parameters including an analyte identifier and optional clonality. In some aspects, the immunoassay reference dataset can include one or more of cell and tissue protein abundance, intensity of bands detected for experimentally determined antibodies, protein turnover and half-life data, protein isoform data, protein compartment localization, post-translational modifications, sequence data, and membrane topology data. In some aspects, the immunoassay reference dataset can include each of these data types. In some aspects, the immunoassay reference dataset can include at least these data types.

[0056] In embodiments, the analyte identifier can refer to the name of the analyte. The name can be a common name, a scientific or molecular name, a user-defined name, a symbol, a code, or other way of referencing and identifying the type of analyte. In embodiments, the clonality can refer to at least one of a monoclonal analyte or a polyclonal analyte.

[0057] At 130, embodiments may further apply one or more machine learning networks to the immunoassay reference data to identify target parameters based on the analyte. An exemplary machine learning method that may be used at step 130, according to one embodiment, is further described with respect to FIG. 2. Once optimal parameters are identified, at 140, embodiments may then determine a recommendation based on the target parameters, the recommendation including a loading concentration of at least one of the analyte and the detection reagent. At 150, the recommendation may be output on a user interface.

[0058] In embodiments, the loading concentration of at least one of the analytes may include analytes contained in the sample and / or further contained in a sample buffer, or contained in a solution, buffer, or mixture, e.g., a mixture with other sample components, including other cellular components, other tissue components, or a mixture that includes one or more salts, buffer-related components, stabilizers, chemicals, preservatives, etc. In embodiments, the loading concentration may relate to the amount of analyte or other input for an immunoassay experiment. The loading concentration may vary based on the type of immunoassay (e.g., immunoblot, flow-based, fluorescent, gel-based, etc.), the desired outcome, or other experimental factors as described herein.

[0059] In embodiments, the analyte may comprise at least one of a protein, a hapten, a hormone, a nucleic acid, a peptide, a modified peptide, or a modified form of any of the foregoing analytes, In embodiments, the modified peptide may be formed from at least one of methylation and acetylation.

[0060] Embodiments of the machine learning models described herein can provide multiple utilities, including, but not limited to, informing the end user via a software interface about target protein loading for protein detection and quantification in Western blot assays, informing the range of detection and providing recommendations regarding antibody concentration, lysate preparation, gel type, and selection of detection reagents for protein target detection, and providing tools for selection and design of multiplexed protein detection in immunoassays to guide experimental design.

[0061] Embodiments may further extract immunoassay training and reference data from the images indicative of the experimental immunoassay data, receive information indicative of a recommendation type, the recommendation type being at least one of a troubleshooting solution or an experimental design, and update target parameters based on the immunoassay parameter set or other recommendation type.

[0062] In embodiments, the optimal target parameters may relate to at least one of the immunoblotting method, such as a Western blot application, the transfer method (i.e., associated with immunoassay techniques, such as wet transfer / electroblotting, semi-dry transfer / electroblotting, and dry transfer / electroblotting), and the type of protein gel. Troubleshooting solutions may identify one or more of the analyte source, antibody, and dilution detection information, as described herein. The dilution detection information may include, but is not limited to, antibody dilution and reagent dilution.

[0063] 2 shows an exemplary flow chart of embodiments and models for providing recommendations and / or predictions of optimal analyte detection parameters. In various embodiments, at 210, the systems and methods can extract relevant immunoassay reference data based on experimental input parameters as described herein. At 220, the embodiments can classify the relevant immunoassay reference data into variables and at 230 apply a statistical model to determine relationships between two or more variables.

[0064] Some embodiments can optionally use the determined relationships between two or more variables to train machine learning networks and programs at 240. Whether or not a training step is performed, the systems and methods can determine relationships between experimental input parameters and corresponding reference analyte detection data, as described herein, at 250. Thus, at 260, embodiments can predict target parameters for detecting an analyte.

[0065] Further embodiments can identify immunoassay reference data associated with a subclass of an analyte, determine one or more subclasses of the analyte of interest, and update target parameters based on the detected subclass of the analyte, wherein the subclass of the analyte is a transmembrane protein, a labile protein, a phosphorylation modification, a glycosylation modification, a post-translational modification, a protein form, a protein isoform, a truncation variant of a protein, or a mutant of a protein.

[0066] It is understood that the immunoassay types, statistical models, experimental input parameters, variables, target parameters, analytes, and recommendations can be customized to encompass a variety of inputs, desired outputs, experimental types, and the like, including but not limited to the examples provided below.

[0067] In various embodiments, the experimental input parameters can include data from multiple assays. In embodiments, the experimental input parameters can include one or more of the following: analyte type (e.g., hapten, hormone, nucleic acid, peptide, modified peptide, or modified form of any of the foregoing analytes), protein type, clonality, cell line, detection reagent, antibody concentration, substrate, substrate sensitivity, detection sensitivity, lysate concentration (such as a lysate of a cell sample or tissue sample), cell line, set of proteins, antibody binding change, blocking data, cell lysate preparation data, protein instability, protein stability, gel parameters, analyte amount range, protein mass range, cell line, detection data, lysate type, lysate loading concentration, protein, protein isoform, fragment of protein or post-translationally modified protein, antibody binding site, antibody clonality, antibody dilution, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection multiplicity, or detection sensitivity.

[0068] The experimental input parameters related to the detection data may further include chemical detection data (e.g., chemical substrates), biochemical detection data (e.g., enzymes), sequence-based detection, amplification-based detection, and / or fluorescence detection data. The experimental input parameters may further include an analyte source, a set of proteins and cell lines, a set of constraints received via a user interface that define the desired experimental parameters. The generated immunoassay parameter set may include determined and / or recommended values ​​for a set of variables, the recommended values ​​being determined to assist in achieving a desired outcome of the immunoassay experiment. In an example, the analyte source may be a protein source or other biochemical or molecular source for the analyte.

[0069] In other embodiments, the experimental input parameters may include at least one detection protein or target analyte, a set of proteins, and a set of constraints received via a user interface, where the target parameters specify a recommended value for at least one constraint in the set of constraints. The set of constraints may further include at least one of available lysates, cell lines, tissue types, detection techniques, antibody clonality, haptens, hormones, modified nucleic acids, peptides, antibody clonality, proteins, antigens, analyte size, protein source, cell lines, tissue types, detection sensitivity, loading concentration, protein abundance, antibody effects (e.g., visualization, immobilization on a surface or medium, binding effects, etc.), membrane effects, blocking effects, extraction effects, protein instability, membrane types, analyte abundance, antibody binding variation, antibody clonality, binding affinity, antibody isoform specificity, scaffold type, detection labels, enzymes, detection multiplicities, or detection sensitivity.

[0070] In embodiments, the experimental input parameters may further include a set of variables received via the user interface, the problem being related to detection and the recommendation including recommended values ​​for the set of variables. The set of experimental input parameters may represent an experimental method, the experimental method being a Western blot method, an immunoblot method, a transfer method, and a method utilizing a protein gel.

[0071] In various embodiments, the immunoassay can include one of many types and designs of immunoassay experiments. An embodiment can utilize, for example, a bead-based immunoassay, such as a multiplex assay, a bead-based immunoassay utilizing a panel, or a bead-based immunoassay utilizing an activated surface panel builder. In other embodiments, the flow-based immunoassay can be a lateral flow immunoassay, a flow assay using colored particles, a competitive assay, or the like.

[0072] The statistical models described herein may include at least one of a cost function, a logistic regression model, a multivariate regression model, a random forest model, a neural network model, and a stochastic gradient model. In embodiments, the machine learning network may apply at least one of a regression model, a decision tree-based training model, and a stochastic gradient descent model.

[0073] Target parameters described herein can include at least one of the following: protein, antigen, analyte size, protein source, cell line, tissue type, detection sensitivity, loading concentration, protein abundance such as protein abundance in a sample, protein abundance in a cell, antibody effect (e.g., visualization, immobilization on a surface or medium, binding effect, etc.), membrane effect (e.g., a measurable effect on or associated with an immunoassay membrane), blocking effect (e.g., a measurable effect due to a blocking agent such as an active blocker, a passive blocker, a specialized blocker, a cross-linking blocker, etc.), extraction effect (e.g., resulting from removal of an interfering protein in a sample), protein instability, membrane type, analyte abundance, lysate, antibody binding change, antibody clonality, binding affinity, antibody isoform specificity, scaffold type (e.g., a protein scaffold associated with one or more of the analytes or antibodies), detection label, enzyme, detection multiplicity or detection sensitivity, type of gel, type of membrane transferred to, transfer method, transfer buffer, wash protocol or wash buffer. The immunoassay reference data may include at least one of Western blot data, multiple Western blot captures, quantitative data, optionally quantitative data representing relative abundance of proteins, categorical data (e.g., data that can be divided into groups), protein detection data, and immunocytometry data, as described herein. The immunocytometry data may further include at least one of analyte localization data or analyte intensity data. Such data may refer to analyte localization data within at least one of a cell, an antibody, or a sample, and analyte intensity data for at least one of a cell, an antibody, or a sample. In exemplary scenarios, the analyte may be secreted from a cell, may be present in an extracellular fluid, may be recombinantly expressed, may be targeted to a non-native location, may be present intracellularly in one or more cellular compartments, may be biochemically fractionated, and may be resolved for biophysical / biochemical properties (such as Western blot or isoelectric focusing).

[0074] The quantitative data can include at least one of analyte abundance estimates and analyte detection data from multiple experiments utilizing one or more of different loading concentrations, different detection reagents, and different antibody affinities.

[0075] In embodiments, the immunoassay reference data may be extracted from images representing experimental immunoassay data.

[0076] The recommendations may include one or more of cell line, tissue, analyte loading range, target protein, detection data, lysate type, lysate loading concentration, protein clonality, antibody type, antibody binding site, antibody clonality, antibody dilution range, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection reagent, detection multiplicity, detection sensitivity, optimal range of loading concentration of at least one of the analytes, clonality, antibody type, detection reagent, immunoassay performance prediction, recommended protein source, optimal cell line, cell source, antibody clonality, detection technique, clonality recommendation, antibody recommendation, analyte recommendation, analyte source recommendation, gel type (e.g., particle gel, agarose gel, or other gel that can be used in an immunoassay experiment), recommended detection reagent, antibody type, antibody loading concentration, protein lysate concentration, optimal cell line, optimal protein source, analyte amount, transfer conditions, validation flags, and predicted analyte location, as discussed herein.

[0077] The detection techniques described herein can be applied to at least one of chemiluminescence, fluorescence, enzyme, and colorimetry. Other embodiments can include a target range of loading concentrations of at least one of the analyte and detection reagent. In embodiments, the analyte can be in sample or purified form.

[0078] Immunoassay performance predictions may include, but are not limited to, providing at least one of the following: recommended analyte, recommended detection reagent, antibody concentration, protein lysate concentration, antibody type, optimal cell line, optimal protein source, analyte amount, migration conditions, validation flags, predicted analyte location (e.g., analyte location relative to a gel or medium in which the immunoassay experiment may be performed, location relative to one or more samples, etc.), protein lysate from cell line source, lysate type, antibody dilution range for optimal detection, antibody dilution based on clonality and host backbone type, type of detection reagent, type of detection technology, and optimal detection technology to ensure linearity.

[0079] In various embodiments, determining a recommendation for detecting multiple proteins in a single cell can be based on optimal parameters, the recommendation including a loading concentration of at least one of an analyte and a detection reagent. The recommendation can further include a type of dye for detecting at least one of the proteins or analytes, the type of dye being at least one of an Ab-conjugate, a secondary Ab conjugate, or a strong dye for detecting low abundance analytes.

[0080] User interfaces, as described herein, can include instrument consoles, web tools, graphical user interfaces, and displays on computing devices.

[0081] FIG. 17 further illustrates an exemplary method 1700 for identifying target parameters of an experiment based on an analyte using one or more machine learning networks as described in FIG. 1 and FIG. 2, according to some embodiments. In some embodiments, multiple machine learning networks (also referred to herein as "models") may be used to determine the target parameters. In step 1702, an analyte of interest is received. In step 1704, a first machine learning model may determine a cell line recommendation for an experiment using the analyte of interest. In some aspects, the first machine learning model may determine a tissue recommendation. Step 1704 is further described by FIG. 18. In step 1706, a second machine learning model may rank the cell line recommendations to determine a selection priority of the cell line. In some aspects, the second machine learning model may rank the tissue recommendations. Step 1706 is further described by FIG. 19. In step 1708, a third machine learning model determines detection agent, loading amount, and antibody dilution recommendations for designing an experiment to be performed on the analyte of interest. Step 1708 is further explained with reference to FIG.

[0082] 18 shows a method for determining cell line recommendations for use in immunoassay experiments using a machine learning model. In one embodiment, this first model may be trained to identify a target parameter based on the analyte. The target parameter may be a cell line or tissue.

[0083] In step 1802, expression data may be retrieved from one or more datasets, such as public RNA sequencing sources (e.g., ARCHS, CCLE, and other public datasets), and a dataset of immunoassay experimental results may be retrieved. In some aspects, the immunoassay experimental result dataset includes results of multiple immunoassay experiments, each experiment performed on an analyte from a plurality of analytes. Each immunoassay experimental result may include the analyte, analyte abundance, clonality, and antibody backbone.

[0084] In step 1804, the expression data and immunoassay experimental results may be cleaned and normalized to prepare the data as immunoassay reference data that may be used in training the first model. In step 1806, the immunoassay reference data may be converted into a feature matrix. In some aspects, the immunoassay reference data may include reference input parameters and reference analyte detection data. In some aspects, the reference input parameters may include transcriptomics (TPM) abundance, clonality, and / or antibody scaffolds. In some aspects, the antibody scaffolds may be derived from one or more mammals, such as mice, rabbits, or mice and rabbits. In some aspects, the reference analyte detection data may identify whether the analyte was detected in the cell line or tissue used in each respective immunoassay experiment.

[0085] In step 1808, a first model can be trained using the feature matrix. The first model can be a machine learning model. In some aspects, the first model can be a logistic regression model. In step 1810, a performance analysis can be performed on the first model. The performance analysis can be performed, for example, using a validation set of experimental input parameters and corresponding reference analyte detection data. Based on the results of the performance analysis, in step 1812, the trained first model can be stored so that it can be accessed and used by the computing device. Once the first model is stored, it can be used to recommend one or more cell lines or tissues in which the analyte of interest is likely to be detected.

[0086] Step 1814 begins the process of using the trained first model to recommend one or more cell lines or tissues in which the analyte of interest is likely to be detected. In step 1814, an identifier of the analyte of interest for an immunoassay experiment may be received. In step 1816, the computing system may retrieve experimental input parameters for the received identifier from a database of analyte data, such as, but not limited to, Uniprot, ARCHS, and / or CCLE. The experimental input parameters may be, for example, TPM abundance, clonality, and antibody scaffolds, such as mouse and / or rabbit antibody scaffolds. In step 1818, the first model may be run on the experimental input parameters. In step 1820, the first model may output a cell line or tissue recommendation for detecting the analyte of interest in the immunoassay experiment. The output may be a list of cell line or tissue recommendations, or may be visually presented by a UI, as described herein.

[0087] 19 illustrates a method for ranking cell line recommendations for use in immunoassay experiments using a machine learning model, according to some embodiments. In one aspect, this second machine learning model may be trained to identify a target parameter based on the analyte. The target parameter may be a ranking order of cell line or tissue recommendations for the immunoassay experiment.

[0088] In step 1902, expression data may be retrieved from one or more datasets, such as public RNA sequencing sources (e.g., ARCHS, CCLE, and other datasets with literature and / or tested antibody data), and target specific features may be retrieved from one or more datasets, such as public analyte databases, such as, but not limited to, Uniprot and / or literature tested antibody data. Output obtained from the first model is also retrieved. In some aspects, the target specific features include data for multiple analytes. The target specific features may include one or more of isoelectric point (pI), transmembrane (TM) domains, TPM abundance, turnover, peptide number, and localization.

[0089] In step 1904, the expression data, target specific features, and output from the first model may be cleaned and normalized to prepare the data into immunoassay reference data that may be used in training the second model. In step 1906, the immunoassay reference data may be converted into a feature matrix. In some aspects, the immunoassay reference data may include reference input parameters and reference analyte detection data. In some aspects, the reference input parameters may include one or more of isoelectric point (pI), transmembrane (TM) domain, TPM abundance, turnover, peptide number, and localization. In some aspects, the reference analyte detection data may be a ranking of cell lines or tissues for detecting the corresponding analyte in an immunoassay experiment.

[0090] In step 1908, a second model can be trained using the feature matrix. The second model can be a machine learning model. In some aspects, the second model can be a random forest classifier model. In step 1910, a performance analysis can be performed on the second model. The performance analysis can be performed using the experimental input parameters and a validation set of corresponding reference analyte detection data. In step 1912, the second model can be retrained based on an analysis of the performance of each feature. In step 1914, the second model can be saved so that it can be accessed and used by the computing device. Once the second model is saved, it can be used to rank one or more cell lines or tissues that can detect the analyte of interest.

[0091] In step 1916, an identifier of an analyte of interest for an immunoassay experiment and a set of cell lines or tissues recommended for use in the immunoassay experiment may be received. In some aspects, the set of cell lines or tissues may be obtained from a first model, such as the output of step 1820 of FIG. 18. In step 1918, the computing system may retrieve experimental input parameters for the received identifier from a database of analyte data, such as, but not limited to, Uniprot, ARCHS, and / or CCLE. The experimental input parameters may include the isoelectric point (pI), transmembrane (TM) domains, TPM abundance, turnover, peptide count, and localization of the analyte of interest. In step 1920, a second model may be run on the experimental input parameters. In step 1922, the second model may output a ranking order of cell line or tissue recommendations for detecting the analyte of interest in the immunoassay experiment. In some aspects, the ranking order may rank the cell lines based on the likelihood that the analyte will be detected in an immunoassay using the cell lines. The ranking may order the cell lines while also ranking the potential as high, medium, or low. In some aspects, the output may be a list of cell line or tissue recommendations, or may be visually presented by a UI as described herein.

[0092] 20 shows a method for determining cell line recommendations for use in immunoassay experiments using a machine learning model according to some embodiments. In one aspect, this third machine learning model may be trained to identify target parameters based on the analyte. The target parameters may be parameters for optimal immunoassay experiments of the analyte.

[0093] In step 2002, expression data may be retrieved from one or more datasets, such as public RNA sequencing sources, such as, but not limited to, ARCHS, CCLE, and other public datasets. A dataset of experimental detection data from immunoassay experiments may also be retrieved. In some aspects, the experimental detection dataset includes the results of multiple immunoassay experiments, each experiment performed on an analyte from multiple analytes. The experimental detection dataset may include, for example, one or more of pixel counts, clonality, dilution factor, loading amount, exposure time, and detection agent for each immunoassay experiment.

[0094] In step 2004, the expression data and the experimental detection dataset may be cleaned and normalized to prepare the data into immunoassay reference data that may be used in training the third model. In step 2006, the immunoassay reference data may be converted into a feature matrix. In some embodiments, the immunoassay reference data may include one or more of pixel counts, clonality, dilution factors, loading doses, exposure times, and detection agents for multiple immunoassay experiments.

[0095] In step 2008, a third model can be trained using the feature matrix. The third model can be a machine learning model. In some aspects, the third model can be a multivariate linear regression model. In some aspects, the third model can use multiple linear regression models, e.g., Atto, Pico, Dura, and ECL. The third model may use 1, 2, 3, 4, 5, or any other number of models. In step 2010, a performance analysis can be performed on the third model. The performance analysis can be performed using a validation set of experimental input parameters. Based on the results of the performance analysis, in step 2012, the third model can be stored so that it can be accessed and used by the computing device. Once the third model is stored, it can be used to recommend experimental parameters for an immunoassay experiment for an analyte of interest.

[0096] In step 2014, an identifier of an analyte of interest for an immunoassay experiment may be received. In step 2016, the computing system may retrieve experimental input parameters for the received identifier from a database of analyte data, such as, but not limited to, Uniprot, ARCHS, and / or CCLE. The experimental input parameters may be one or more of a loading fraction, exposure time, clonality, and dilution factor. In step 2018, a third model may be run on the experimental input parameters. In step 2020, the third model may output a set of recommendations for an optimal immunoassay experiment. For example, the third model may output one or more of a recommended detection agent, loading amount, and antibody dilution for an immunoassay experiment for the analyte of interest. The output may be a list of recommendations or may be visually presented by a UI, as described herein.

[0097] Figure 3 shows the relationship between protein, detection agent affinity, and detection reagent sensitivity according to embodiments described herein. For immunoassay experiments, many factors can affect the output and successful observation. Balancing variables and input factors for immunoassay experiments is a consideration for effective data collection and analysis.

[0098] FIG. 3 highlights three important factors that influence performance data and which embodiments of the disclosed technology consider its optimization and recommendation techniques, such as identification of target parameters and immunoassay parameter sets. The amount of protein available for detection, detection reagent affinity, and detection reagent affinity each affect the effectiveness of the other variables. Protein available for detection can be provided via reference data as described herein for an experiment. Absolute quantification based on transcripts per million / protein intensity-based absolute quantification measured in Fmol / cell. Protein-specific quantitation (TPM / iBAQ) is just one example of protein-specific reference data that can be used for detection.

[0099] The affinity of the detection reagent can be obtained and / or provided by training data, e.g., training data for machine learning operations. The affinity of the detection agent can depend, for example, on clonality, e.g., monoclonal or polyclonal, and can be provided using one or more organisms, e.g., rabbit or mouse. K and K' can indicate the variance associated with the sensitivity of the detection reagent. The detection reagent sensitivity can also be obtained and / or provided by training data, e.g., ECL, Pico Plus, SuperSignal, and Atto. As a result, the sensitivity of the detection reagent can affect and be affected by the proteins available for detection, and the relationship between the two can be characterized, in an example, by Z=(Var).

[0100] Additional factors that may affect optimal immunoassay data include, but are not limited to, secondary antibody effects, membrane effects, blocking effects, protein instability, and extraction effects. For example, unstable proteins may result in poor extraction quality. Data cleansing filters can be applied to the data points to also eliminate the effects of outliers.

[0101] Figures 4 and 5 show graphical representations of a conceptual framework for immunoassays. The figures show that dose-response curves can be modeled as sigmoidal functions. Figure 4 shows an example dose response for antibody clonality and TPM values. Figure 5 shows a sigmoidal dose-response curve overlaid with a corresponding linear model. A linear model of dose response can be represented by the standard linear form, y=b0+b1x, and a logistic model can be represented by p=1 / (1+e^-(b0+b1x)). These linear and logistic models can be implemented with one or more machine learning algorithms discussed herein.

[0102] FIG. 6 shows an exemplary flow chart for generating a model and providing immunoassay parameter sets, target parameters, and other recommendations and predictions according to embodiments described herein. To generate a model, data preparation 610 can include input on data detection agents such as TPM / iBAQ, clonality information, etc. A query 640 for TPM / iBAQ values ​​can be performed in conjunction with a check 620 of an outlier protein list. In an example, TPM and iBAQ do not need to be correlated. If the check is positive, an outlier protein is identified to be present and a filter / count prediction can be output at 630. If the check is negative, indicating that no outlier protein and correlation exists, input matrix preparation 650 and logistic model operation 670 are performed.

[0103] The input matrix 650 can assist in the training 660 of the logistic regression model and utilize the information from the logistic model executed at 670. Based on the logistic regression model training 660, the model parameter significance can be output at 675. Additionally, the logistic model executed at 670 can assist in determining the probability of detection at 680. The detection probability helps in determining a possible shift to a preferred reagent and / or linear range. In one example, when the probability is 0.6 < P < 0.85, a preferred agent 695 such as the cell line ECL / SS / Atto is provided. When P < 0.6 or P > 0.85, the model prediction shifts to the linear range at 690. The endpoints 0.6, 0.85 can optionally be associated with any option and, in the example, can be defined and / or selected by the user.

[0104] Figure 7 shows an example of an experimental design workflow for multiplex protein detection. The experiment can initially utilize a set of proteins and an optional cell line for the input at 720. In this example, the set of proteins can include five proteins (A - E). Multiple experimental data and reference data can provide additional information for detection recommendations and probability determination. In the example, cell - tissue and protein abundance reference data 730 and / or antibody data and clone information 740 include the reference data for the model described herein. Embodiments can generate a graphical representation 745 of the predicted output for each of proteins A - E. The graphical representation and associated data can provide recommendations 750 and predictions 760 for one or more experimental variables and / or outputs.

[0105] In the example, the predictions and recommendations can include immunoassay parameter sets, probability of detection, threshold loading values, target values for linear detection, and missing proteins. Further, the output can provide a recommended antibody, a suggested dilution for one or more products, and a recommended detection agent.

[0106] 8-9 illustrate user interfaces associated with embodiments described herein. The embodiments may be provided on one or more computing systems, displays, web tools, applications, and the like. FIG. 8 illustrates an initial screen with an option for a user to select the details of an experiment to be evaluated. In the illustrated example, the user may select the type of protein (e.g., coilin) ​​and clonality (e.g., mono, poly). It should be understood that the details provided on the user interface need not be limited to the depicted categories, but may be tailored based on any of a number of factors, including, but not limited to, experiment type, desired outcome, typical variables, variables of particular interest, and the like.

[0107] The UI may be utilized to implement a method such as that described by Figure 1. In one aspect, the UI may facilitate the design of an experiment, specifically a Western Blot experiment, so that the experiment is optimally performed with little or no testing. The UI may be hosted by a host server and may be connected to the Internet. In other aspects, the UI may not be connected to the Internet, but may instead reflect operations that are performed locally.

[0108] In one aspect, the UI may prompt the user to design a new experiment. If the user selects to design a new experiment, the module provides recommendations for reagents that may be used to detect an analyte of interest, such as a protein of interest. In other aspects, implementations of the UI may be used to provide recommendations for various analytes as described herein.

[0109] Figure 9 shows an exemplary dataset output for a selected protein, which may be input on an initial screen such as, for example, Figure 8. The dataset output may provide information regarding the use of one or more of cell lines, TPM, ECL, and SuperSignal. Thus, the output may inform the user of one or more output predictions that may be used to aid in experimental design and identify target variables and conditions for a desired experiment.

[0110] FIG. 21 shows an alternative version of FIG. 8. As shown in FIG. 21, the UI may prompt the user to input a protein. The input may be selected from a menu or may be entered in free text. The protein may be entered using, for example, its name, Uniprot ID, alias, gene name, and / or gene synonym. The UI may present a list of protein suggestions as a lookup table from which the user may select a protein. In some aspects, the user may enter multiple proteins of interest.

[0111] Once the UI receives the protein, the user may be prompted to select whether a cell line will be used to complete the experiment. In some embodiments, the user may select that instead of a cell line, a tissue will be used for the experiment.

[0112] The UI may also prompt the user to optionally input a cell line selection. If a cell line selection is input, the model may determine whether the protein can be detected from any of the cell lines in the cell line selection. If no cell line is input, the model determines whether the protein can be detected from any of the cell lines in the complete list of available cell lines.

[0113] The user may choose to receive protein results, which may include one or more of data about protein annotations, cell lines, detection reagents, and antibodies.

[0114] The protein annotation results can display information about the protein, which may include, but is not limited to, the protein's name, Uniprot ID, gene name, mass, post-translational modifications, and / or isoelectric point (pI).

[0115] As shown in FIG. 21, the cell line results can provide recommendations for cell lines that are most likely to detect the protein if the experiment is performed. The cell line results can report the determined optimal cell lines obtained from the first model, as described by FIG. 18. The UI may display the cell line recommendations along with a pictorial representation of the cell lines sorted by lineage and probability of detection, for example, as illustrated in FIG. 21. Each cell line in the pictorial representation may be indicated by a mark, for example, a dot. The user can hover over each mark to display the cell line. In some aspects, the cell culture conditions may also be displayed when hovering over a mark.

[0116] If the user inputs multiple proteins, the cell line results can display cell line recommendations in a matrix with the cell line on one axis and the protein on the other axis. The matrix displays "yes" or "no" depending on whether the protein can be detected in the corresponding cell line. In some aspects, the user can hover over a cell line and the cell culture conditions and lineage of the cell line can be displayed.

[0117] As shown in Figure 22, after receiving the cell line results, the user can select a cell line from the displayed cell lines. The cell line selection can be input into a second model, which ranks the selected cell lines based on the likelihood of the protein being detected in the cell line, as described by Figure 19. In some embodiments, the UI can display the ranking and classify the cell lines as "high," "medium," or "low" with respect to potential detection.

[0118] As illustrated in FIG. 23, an exemplary lysate and detection reagent result may display an experimental recommendation for the lysate type, gel type, and house-keeping protein (HKP) to be used for the protein. The recommendation may be determined from a data lookup model that utilizes known protein mass to match the protein mass to the recommended lysate type, gel type, and HKP. In some aspects, the lysate type may be displayed along with the number of proteins for which it may be used. In some aspects, the gel type is displayed along with the proteins that may be detected using the gel type. In some aspects, the HKP may be displayed as the number of recommended HKPs, e.g., 23 recommended HKPs may be displayed as "23 HKPs." The user may select a numbered HKP recommendation and the UI may display a list of the recommended HKPs along with the name, mass, clonality, and / or backbone of each HKP along with a corresponding SKU link for an antibody that may be used to detect the HKP.

[0119] Similarly, exemplary antibody results may be displayed with their names and SKU links. In some aspects, a link may be available on the results page that may direct the user to a web page detailing all available primary antibodies.

[0120] The UI may display one or more recommendations for designing an experiment with the protein, as illustrated in FIG. 24. The UI may visually display results from the third model, as described by FIG. 21. These results include recommended loading amounts, antibody dilutions, and detection reagents for detecting the protein of interest. In some embodiments, the pixel density for the experiment may be displayed as "high," "medium," or "low."

[0121] Figure 10 shows an exemplary scenario related to parameter optimization and recommendation. In the illustrated example, experiment 1010 and experiment 1020 show output differences related to increasing protein loading. In experiment 1010, the lysate concentration is 30 μg, and experiment 1020 increases the lysate concentration to 40 μg. In both experiments, the primary dilution is 1:1000 and the detection reagent is Atto. Therefore, the difference in results can be attributed to the difference in lysate concentration.

[0122] In this example and other common user scenarios, low detection results (i.e., experiment 1010) can often be attributed to one or more of low protein abundance, extraction losses, lower antibody affinity based, excessive blocking, and weaker detection agents. In such scenarios, low detection can be addressed by one or more of the following actions: loading additional protein, reducing blocking, modifying the protein preparation, changing the membrane type, using a more sensitive detection agent, and replacing the antibody. Each of the above factors may have a slightly different impact on the experiment, and the number of tests / experiments a user conducts increases dramatically because the user must decide which factors to adjust and test. For example, if there are six factors that affect detection and there are three to four possible user adjustments per factor, the result is 24 to 30 user tests to test factors and identify the factor that optimizes detection. Thus, applying the machine learning models described herein can efficiently test and predict experimental results for multiple factors and recommend ideal factors for experimental use.

[0123] FIG. 11 illustrates another common user scenario where target signal-to-noise ratios require identification. In various examples, optimizing the signal-to-noise ratio can improve data from unacceptably noisy to publishable. Furthermore, such improvements can improve assay performance by identifying target protein-to-antibody ratios. In the illustrated experiments of FIG. 11, experiment 1110 utilizes a primary dilution of 0.5 μg / ml and experiment 1120 utilizes a primary dilution of 0.25 μg / ml. The lysate concentration and detection reagents remained the same between experiments. As a result of the dilution change, experiment 1120 resulted in a cleaner immunoassay in which the markers could be clearly identified.

[0124] In these scenarios, poor signal-to-noise ratios can result from multiple factors, including, but not limited to, suboptimal protein loading, degraded preparations, insufficient antibody volume, excess antibody volume, and suboptimal blocking. User responses to address these issues often include one or more of increasing protein loading, increasing blocking, modifying the protein preparation, changing membrane type, diluting the antibody, and replacing the antibody. The machine learning models described herein can help test these factors and predict experimental recommendations based on previous data and model input information.

[0125] Figure 12 shows a set of experiments utilizing multiple antibodies with different sensitivities. Figure 12 shows how experimental results can change dramatically based on sensitivity levels. Such information, including experimental data from each example, can provide useful reference data for training one or more machine learning models according to embodiments discussed herein. The experiments shown demonstrate the effect of sensitivity differences, which can provide useful information for experiments utilizing similar antibodies and targets.

[0126] Figure 13 shows another set of experiments with different results depending on the change in variables. In the example shown, increasing the lysate load from 30 μg to 50 μg provides the amount required for detection. Again, such data can provide useful information for machine learning models and prediction of the effect on increasing lysate load.

[0127] FIG. 14 shows another example experiment. In this scenario, protein quantification can be improved by decreasing the protein concentration. Antibody titration can be performed in the example to aid in adjusting the antibody concentration and ultimately the protein quantification. Experiments 1410 and 1420 show the improved effect of decreasing the antibody concentration. In experiment 1410, the Ab concentration is decreased from 1 μg / ml to 0.5 μg / ml, improving the clarity of detection. In experiment 1420, the Ab concentration is decreased from 1 μg / ml to 0.5 μg / ml between the first two tests and the last two tests, and the lysate concentration remains at 30 μg for the first two tests and decreases to 20 μg for the last test. The initial decrease in Ab concentration between test 1 and test 2 provides a clear detection result, and the subsequent decrease in lysate concentration between test 2 and test 3 further clarifies the detection result. As with previous experiments described herein, the experimental results provided in FIG. 14 can provide reference data for machine learning models to aid in experimental predictions and recommendations.

[0128] FIG. 15 illustrates the experimental differences that can occur with enhanced reagent sensitivity. In experiment 1510, the first test is developed with ECL for 5 minutes followed by SuperSignal for 2 minutes. The second test is developed with Atto reagent for 5 seconds. The second test gave a clear detection result. In experiments 1520, 1530, the first test was developed with ECL for 3 minutes and the second test was developed with Atto reagent for 2 seconds. In both cases, the change in reagent increases detection.

[0129] troubleshooting It is expected that analytes will migrate through a gel (or other membrane, depending on the type of electrophoresis used) based on a function of mass (molecular weight), such that analytes of different masses can be identified as distinct bands in an immunoblot. Such techniques can be used, for example, to identify and quantify proteins during protein preparation. However, there are many factors that can cause an analyte not to migrate as expected. For example, the charge of the analyte, post-translational modifications (e.g., glycosylation, lipidation), gel type, and buffer type can all affect the migration of the analyte. This difference in migration is referred to herein as a "shift" - a physical shift between the observed migration and the actual (i.e., expected) molecular weight for a given analyte. This shift can affect the outcome of an immunoassay and can lead to errors or misinterpretations in the immunoassay results.

[0130] Therefore, there is a need to develop a method that can determine the migration shift of an analyte in an immunoassay so that the results of the immunoassay are still accurate and usable. Because there are multiple factors that affect migration, and some of these factors are not linear, linear calculations may not be effective in determining the shift. Thus, according to aspects described herein, neural networks can be used to determine the extent to which an analyte has migrated or shifted in an immunoassay experiment.

[0131] FIG. 25 illustrates a method 2500 according to some embodiments. For example, method 2500 can be used to determine an analyte shift in an immunoassay. An analyte is a specimen extracted for analysis from a portion or sample using a particular extraction protocol, and is typically a substance of interest that requires detection. Often, analytes are proteins, but they can be other kinds of molecules of different sizes and types, as long as a method is available to tag the analyte (e.g., staining or antibody detection). In some embodiments, the analytes described herein can be proteins, haptens, hormones, nucleic acids, peptides, modified peptides, or modified forms of any of the aforementioned analytes.

[0132] In step 2502, the computer system can receive an immunoassay dataset having known information about a plurality of known analytes. The immunoassay dataset may be received from a plurality of sources, including, but not limited to, a database of analyte data, a researcher, or an organization. The immunoassay dataset can include data collected from immunoassay experiments of a plurality of analytes. In some aspects, the immunoassay experiment can be a bead-based immunoassay, such as a multiplex assay, a bead-based immunoassay utilizing a panel, or a bead-based immunoassay utilizing an activated surface panel builder. In other aspects, the flow-based immunoassay can be a lateral flow immunoassay, a flow assay using colored particles, a competitive assay, or the like. In some aspects, the immunoassay can be a single or multiplex Western blot, or an SDS-PAGE (sodium dodecyl sulfate-polyacrylamide gel electrophoresis) gel.

[0133] The immunoassay dataset may include a set of input parameters that may correspond to the degree of shift of the analyte in the gel. In some aspects, the set of input parameters may include various features including, but not limited to, post-translational modifications such as phosphorylation, glycosylation, ubiquitination, nitrosylation, methylation, acetylation, lipidation, and proteolysis, interchain or polymer crosslinks, disulfide groups, modified residues, isoelectric point (pI), gel type, and buffer type, to name a few, each of which corresponds to an immunoassay for a particular analyte. Each set of input parameters corresponds to the degree of shift of the analyte. The immunoassay dataset may have variations in reagents, cell lines, analytes, antibodies, gel types, buffer types, or any other variations of the aforementioned features, from which different attributes that contribute to the molecular weight shift may be identified by the neural network. In some aspects, the immunoassay dataset may include negative data.

[0134] In step 2504, a machine learning process can be used to determine the degree of band shift that a particular analyte undergoes in the immunoassay. Machine learning involves the development and use of computer systems that can learn and adapt without following explicit instructions by using algorithms and statistical models to analyze patterns in data and draw inferences therefrom. Machine learning models suitable for the disclosed embodiments may include, for example, but not limited to, supervised learning, semi-supervised learning, unsupervised learning, or augmented models. A supervised learning model may be trained on the labeled data, and exemplary conditions associated with the desired output are fed to the machine learning model during training. Some non-limiting examples of supervised learning models include, for example, but not limited to, nearest neighbor, naive Bayes, decision trees, support vector machines, neural networks, or any machine learning algorithm suitable for image analysis and / or ranking problems. In some embodiments, the machine learning process may include a feed-forward non-deep neural network or a deep learning network. For the purposes of this disclosure, a non-deep feed-forward network will simply be referred to as a neural network. The neural network can be trained using an immunoassay data set to determine the degree of band shift that a particular analyte experiences in the immunoassay. In some embodiments, the neural network is developed on a computer system including a memory and a processor. The neural network can be constructed using any neural network architecture, such as an unsupervised pre-trained network, a convolutional neural network, a recurrent neural network, a recurrent neural network, etc. In some embodiments, the neural network has at least two hidden layers.

[0135] The neural network may be trained using an immunoassay data set, and input parameters, also known as features, and the corresponding degree of analyte shift may be used as input. In some aspects, the input parameters may include glycosylation of the analyte. In some aspects, the input parameters may include glycosylation, disulfide bonds, modified residues, MOPS (3-(N-morpholino)propanesulfonic acid), and ubiquitination. In some aspects, the input parameters may include glycosylation, disulfide bonds, modified residues, MOPS (3-(N-morpholino)propanesulfonic acid), ubiquitination, lipidation, MES (2-(N-morpholino)ethanesulfonic acid), isoelectric point (pI), 4-12% gel type, 10% gel type, SUMOylation, Tris acetate, 3-8% gel type, 12% gel type, and crosslinking. In some aspects, the input parameters may include additional or fewer parameters than those listed. In some embodiments, input parameters may include polymer modifications, charge-affecting features, and gel type. Features may be extracted from existing databases (e.g., Uniprot). A neural network may be trained to output the degree of analyte shift in the gel.

[0136] In step 2506, once the neural network is trained, the neural network may analyze experimental data of the analyte of interest for an immunoassay experiment. An immunoassay experiment may be completed for detection of the analyte of interest, and the experimental data may be input to the neural network. The experimental data may include data about the analyte of interest, as well as specific experimental parameters of the immunoassay experiment, including, but not limited to, post-translational modifications, glycosylation, lipidation, disulfide groups, modified residues, and isoelectric point for the analyte of interest, and gel type and buffer type for the immunoassay experiment.

[0137] In step 2508, once the analysis is completed, the degree of shift for the band containing the analyte of interest in the immunoassay experiment is determined based on the output of the neural network. The shift is determined by the degree to which the band is predicted to shift according to the results from the analysis by the neural network. In some embodiments, the band may be observed in the immunoassay experiment and contains the analyte of interest.

[0138] In step 2510, in some embodiments, the immunoassay images taken from the immunoassay experiment may be analyzed and marked. The analysis may mark frames, lanes, and bands of the immunoassay image. The analysis may be performed by analysis software (e.g., iBright™).

[0139] In step 2512, the extent of band shift may be marked on the immunoassay image and displayed for a user to visually observe the extent to which the band of the analyte of interest has shifted.

[0140] Once the extent of the shift has been determined, the neural network may link the cause of the extent of the shift to one or more of the experimental parameters based on the immunoassay data set on which the neural network was trained. The neural network can then be used to identify changes to the experimental parameters and optimize the parameters of the experiment for the sample and analyte of interest.

[0141] FIG. 26 is an exemplary diagram of an immunoassay experiment 2600. In some embodiments, the immunoassay experiment 2600 can be a Western blot experiment. Each number on the x-axis represents a lane, and each letter on the y-axis represents a molecular weight / mass. In this example, the band of interest is the band located at the degree of shift determined by the neural network in step 2508. Frame 2602 constitutes the range of the entire immunoassay experiment. In lane 1, band 2604 is the band of interest. In lane 1, there may also be band 2606, which is a non-specific band. A non-specific band is determined to be a band of analyte that is marked by the detection reagent but does not contain the analyte of interest. Thus, it is reported that the band of interest was found along with the non-specific band. In lane 2, band 2608 is also a band of interest. Thus, it is reported that the band of interest was found in lane 2. Lane 3 does not have a band, and therefore lane 3 is reported to be a blank lane. In lane 4, band 2610 is not where the band of interest was determined to be present. Thus, lane 4 is reported as having no band of interest, but containing non-specific bands. In certain cases where the analyte may exhibit post-translational modification (PTM) in a native cell or tissue context, or under certain treatment conditions, or in a cell cycle stage specific manner, multiple bands may be encountered. In such cases, multiple bands within a defined range with differential shifts in mass, depending on the number of sites available for PTM, are reported as bands of interest. Multiple bands can be seen by the exemplary immunoblot shown in Figure 31.

[0142] FIG. 27 illustrates a method 2700 for a system that may be used to determine the extent of analyte band shift that has occurred, according to some embodiments. In some embodiments, the method 2700 may be implemented by a computer system having a memory and a processor. In some embodiments, a user interface (UI), e.g., a website, application, data / content source, etc., with which a user may interact, may facilitate the method 2700 to determine and display analyte band shifts in an immunoassay, such as a Western blot experiment, so that the experiment is optimally performed. The UI may be hosted by a host server and may be connected to the Internet. In other embodiments, the UI may not be connected to the Internet. In some embodiments, the UI may be implemented with existing software, e.g., iBright™, Thermo Fisher Connect Platform™, etc.

[0143] In some embodiments, method 2700 may be initiated by a user to troubleshoot experimental results that are suboptimal. For example, method 2700 may be initiated when the results of an immunoassay analysis do not include or match the expected results (e.g., when the bands detected on a Western blot are different from the bands expected for the analyte of interest). Method 2700 may also be initiated by a user to quality check the accuracy of the system even if the output is not suspected to be inaccurate.

[0144] In step 2702, the system may receive experimental data from a user. The experimental data includes an identifier of an analyte of interest and an immunoassay image. In some aspects, the experimental data may also include, for example, but not limited to, one or more of an identifier of a cell line, an identifier of a molecular marker, a lysate type, a loading concentration, or a gel type used in the immunoassay experiment. In some aspects, the cell line may be a tissue lysate, a recombinant protein, or a synthetic protein.

[0145] An analyte identifier may refer to the name of the analyte, which may be, for example, but not limited to, a common name, a scientific or molecular name, a user-defined name, a symbol, a code, or other way of referencing and identifying the type of analyte.

[0146] The immunoassay image may be an image of an immunoassay experiment performed to detect an analyte corresponding to the identifier. In some embodiments, the immunoassay image may be an image of a bead-based or flow-based immunoassay experiment. In some embodiments, the immunoassay image may be an image of a Western blot experiment. In another embodiment, the immunoassay image may be a stained image of an SDS-PAGE gel. The immunoassay image may have features including bands, lanes, and frames. A band is the location where an analyte migrates in an immunoassay. In some embodiments, there may be multiple bands in one immunoassay image. A lane is a particular panel of an immunoassay image where one experiment has been completed. In some embodiments, an immunoassay image may have multiple lanes when there are multiple experiments completed for the same immunoassay. A frame includes all the lanes and bands in the immunoassay image.

[0147] FIG. 28 is an example of a UI that may be used when performing step 2702, according to some embodiments. In some embodiments, the UI may be included in a UI that utilizes the method of FIG. 1. A pop-up window may be displayed that allows the user to browse and select a file, such as an immunoassay image, for the experiment. In some embodiments, the immunoassay image may be dragged and dropped into the pop-up window, or the file may be selected from a list of accessible files. Once the user selects the file, the file is uploaded. Once the upload is complete, a "troubleshoot" workflow may be launched.

[0148] 29A and 29B are examples of UI pages that may be displayed after an immunoassay image has been uploaded, according to some embodiments. The UI may prompt the user to enter experimental data for the immunoassay experiment. For example, but not by way of limitation, the experimental data may include an identifier for the analyte used, a molecular marker used, the type of gel used, the cell line used in each lane, the loading concentration in each lane, and / or the type of lysate used in each lane for the immunoassay experiment. In some embodiments, the user may be able to adjust the analysis of the image, such as by editing one or more of the frames, lanes, or bands. In some embodiments, the UI may present a series of questions for the user to answer to help guide the troubleshooting process. The questions may relate to the experimental data, for example, but not by way of limitation, as discussed above.

[0149] Returning to Figure 27, in step 2704, the immunoassay image may be analyzed to mark features within the immunoassay image. This may include any bands, lanes, or frames found in the immunoassay image. In some aspects, the analysis may be performed by analysis software. In some aspects, the UI may provide the user with the ability to adjust the results of the analysis by editing the frames, lanes, and / or bands of the image.

[0150] In step 2706, a neural network as described in step 2504 may be used to determine the degree of shift that has occurred to a band on the immunoassay image. The band may be a band containing an analyte of interest that corresponds to the identifier received in step 2702. The experimental data received from the user in step 2702 may be input to the neural network for analysis. The identifier may be used by the system to retrieve data regarding the analyte that is also input to the neural network. The degree of shift of the band is then determined. If there are multiple lanes, each lane may be analyzed separately by the neural network to determine the shift of the band in each lane.

[0151] In step 2708, the extent of the band shift is displayed on the immunoassay image. In some embodiments, the shift may be displayed by the UI. Figures 30A and 30B are an example of a UI page that is displayed if a shift is determined. If the band of interest is not detected, the UI page can display information about what may have gone wrong in the experiment.

[0152] In some embodiments, the observations of each lane are provided by the UI based on the results of the neural network. In some embodiments, the UI can display the band shift for each lane, whether or not the band of interest was detected. In some embodiments, the band of interest can be highlighted in each lane on the immunoassay image. In some embodiments, the observations of all lanes can be displayed to the user via the UI, or a subset of lanes can be displayed to the user based on receipt of filtering parameters. If there is a problem with the experiment, for example, if the gel is used incorrectly, or if the cell line is not suitable for the answer sought, accompanying data and / or details can be provided to the user. For example, for each lane, the user can see one of the following conclusions: (1) the band of interest was found, (2) the band of interest was found along with a non-specific band, (3) the band of interest was not found, and (4) no band was found, so the lane was blank.

[0153] In some embodiments, the molecular weight of each band based on the experimental data of molecular markers is calculated by computer system.The molecular weight can then be displayed in a data table.Figure 30B shows an example of a UI page that displays a data table with observations.

[0154] The system may also determine whether and / or how the experimental data should be corrected for each lane depending on the results of the immunoassay images and analysis. In some embodiments, the correction may be determined by a relationship between the experimental data and known analyte detection data. In some embodiments, the correction may include a suggestion of whether the gel type or cell line used in each lane is suitable for detecting the analyte of interest. In some embodiments, the UI may display the determined correction.

[0155] In some aspects, a user may be able to select a particular band to obtain further details regarding the results of the troubleshooting analysis.

[0156] Based on the band shifts, results, and band conclusions of the experiment, the system can initiate an experiment planning module based on the results of the troubleshooting analysis to optimize the immunoassay experiment and provide a plan for detection of the analyte of interest. For example, using a predictive model, the plan may suggest differences in experimental settings such as exposure time and / or amount of lysate that needs to be loaded to optimize the level of analyte detection within a given range.

[0157] FIG. 16 illustrates an exemplary computing environment 1600 suitable for implementing aspects of an embodiment of the disclosed technology, including a control system that can integrate one or more devices, computing, and lighting systems. As used herein, the phrase "computing system" generally refers to a dedicated computing device having processing power and storage memory that supports software, applications, and operating software underlying the execution of computer programs. As used herein, an application is a dedicated program with a small storage size that is downloaded to a computing system or device. As shown in FIG. 16, the computing environment 1600 includes a bus 1610 that directly or indirectly couples the following components: a memory 1620, one or more processors 1630, an I / O interface 1640, and a network interface 1650. The bus 1610 is configured to communicate, transmit, and transfer data, control, and commands between the various components of the computing environment 1600.

[0158] The computing environment 1600 typically includes a variety of computer readable media. Computer readable media can be any available media that can be accessed by the computing environment 1600 and includes both volatile and nonvolatile media, removable and non-removable media. Computer readable media can include both computer storage media and communication media. Computer storage media does not include, and in fact explicitly excludes, signals themselves.

[0159] Computer storage media includes volatile and nonvolatile, removable and non-removable, tangible and non-transitory media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules, or other data. Computer storage media may be RAM, ROM, EE-PROM, flash memory or other memory technology, CD-ROM, DVD or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium or computer storage device that can be used to store the desired information and that can be accessed by the computing environment 1600.

[0160] Communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media. The term "modulated data signal" means a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. By way of example, communication media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, RF, infrared and other wireless media. Combinations of any of the above are also intended to be included within the scope of computer readable media.

[0161] The memory 1620 includes computer storage media in the form of volatile and / or nonvolatile memory. The memory may be removable, non-removable, or a combination thereof. The memory 1620 may be implemented using hardware devices such as solid-state memory, a hard drive, an optical disk drive, etc. The computing environment 1600 also includes one or more processors 1630 that read data from various entities such as the memory 1620, an I / O interface 1640, and a network interface 1650.

[0162] The I / O interface 1640 allows the computing environment 1600 to communicate with different input and output devices. Examples of input devices include keyboards, pointing devices, touchpads, touch screens, scanners, microphones, joysticks, etc. Examples of output devices include display devices, audio devices (e.g., speakers), printers, etc. These and other I / O devices are often connected to the processor 1610 through a serial port interface coupled to the system bus, but may also be connected by other interfaces such as a parallel port, game port, or universal serial bus (USB). A display device may also be connected to the system bus through an interface such as a video adapter that may be part of or connected to a graphics processor unit. The I / O interface 1640 is configured to coordinate I / O traffic between the memory 1620, the one or more processors 1630, the network interface 1650, and any combination of input devices and / or output devices.

[0163] The network interface 1650 enables the computing environment 1600 to exchange data with other computing devices over any suitable network. In a networked environment, program modules depicted relative to the computing environment 1600, or portions thereof, may be stored in remote memory storage devices accessible via the network interface 1650. It will be appreciated that the network connections shown are exemplary and other means of establishing a communications link between the computers may be used.

[0164] By way of example and not limitation, a cloud computing system can be used to implement aspects of the disclosed subject matter. Cloud-based computing generally refers to a networked computer architecture in which application execution, service provisioning, and data storage can be divided to some extent between clients and cloud computing devices. A "cloud" can refer to a service or a collection of services accessible over a network, e.g., over the Internet, by, for example, clients, server devices, and by other cloud computing systems.

[0165] In one example, multiple computing devices connected to the cloud can access and use a common pool of computing power, services, applications, storage, and files. Thus, cloud computing enables a shared pool of configurable computing resources, e.g., networks, servers, storage, applications, and services, that can be provisioned and released with minimal administrative effort or coordination by a cloud service provider.

[0166] As an example, a cloud-based application may store copies of data and / or executable program code within a cloud computing system and allow client devices to download at least a portion of this data and program code as needed for execution on the client device. In some examples, the downloaded data and program code may be tailored to the capabilities of the particular client device, e.g., a personal computer, tablet computer, mobile phone, and / or smartphone, accessing the cloud-based application. Additionally, splitting application execution and storage between the client device and the cloud computing system may allow more processing to be performed by the cloud computing system, thereby leveraging, for example, the processing power and capabilities of the cloud computing system.

[0167] Cloud-based computing may also refer to a distributed computing architecture in which data and program code for a cloud-based application is shared on a near real-time basis among one or more client devices and / or cloud computing devices. Portions of this data and program code may be dynamically distributed as needed or otherwise to various clients accessing the cloud-based application. Details of the cloud-based computing architecture may be largely transparent to users of the client devices. By way of example and not limitation, a PC user device accessing a cloud-based application may not be aware, for example, that the PC is downloading program logic and / or data from a cloud computing system or that the PC is offloading processing or storage functions to the cloud computing system.

[0168] A cloud platform may include a client interface front end for a cloud computing system. Such an architecture may represent a queue for processing requests from one or more client devices. A cloud platform may be coupled to cloud services to perform functions for interacting with client devices. A cloud infrastructure may include the services, logging, analytics, and other operational and infrastructure components of a cloud computing system. A cloud knowledge base may be configured to store data for use by a network such that the cloud knowledge base may be accessed by any of the cloud services, platform, and / or infrastructure components.

[0169] Many different types of client devices, such as a user's device, may be configured to communicate with components of a network for purposes of accessing data and executing applications provided by one or more processors and computing systems. As described herein, any type of computing device, such as a PC, laptop computer, tablet computer, etc., as well as any type of mobile device, such as a laptop, smartphone, mobile phone, cellular phone, tablet computer, etc., may be configured to transmit and / or receive data to and / or from the network.

[0170] The communication link between the client device and the network may include a wired connection, such as a serial or parallel bus, Ethernet, optical connection, or other type of wired connection. The communication link may also be a wireless link, such as Bluetooth, IEEE 802.11 (IEEE 802.11 may refer to IEEE 802.11-2007, IEEE 802.11 n-2009, or any other IEEE 802.11 revision), CDMA, 3G, GSM, WiMAX, or other wireless data communication link.

[0171] In another example, the client device may be configured to communicate with the network 100 through a wireless access point. The access point may take a variety of forms. For example, the access point may take the form of a wireless access point (WAP) or a wireless router. As another example, if the client device connects using a cellular air interface protocol such as CDMA, GSM, 3G, or 4G, the access point may be a base station in a cellular network that provides Internet connectivity via the cellular network.

[0172] Thus, the client device may include a wired or wireless network interface connection that allows the client device to connect to the network 100 directly or through an access point. By way of example, the client device may be configured to use one or more protocols such as 802.11, 802.16 (WiMAX), LTE, GSM, GPRS, CDMA, EV-DO, and / or HSPDA, among others. Additionally, the client device may be configured to use multiple wired and / or wireless protocols, such as cellular communication protocols, e.g., "3G" or "4G" data connectivity using CDMA, GSM, or WiMAX, and "Wi-Fi" connectivity using 802.11. Other types of communication interfaces and protocols may also be used.

[0173] The above aspects of the disclosure have been described with reference to specific examples and embodiments that are intended to illustrate, not limit, the disclosure. It will be understood that the subject matter presented herein may be implemented as a computer process, a computer controller or computing system, or an article of manufacture such as a computer-readable storage medium.

[0174] Those skilled in the art will also appreciate that the subject matter discussed herein may be practiced on or in conjunction with other computer system configurations other than those discussed herein, including multiprocessor systems, microprocessor-based or programmable consumer electronics, minicomputers, mainframe computers, handheld computers, personal digital assistants, e-readers, cellular telephone devices, biometric devices, mobile computing devices, special purpose hardware devices, network appliances, and the like. The embodiments described herein may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.

[0175] A number of different types of computing devices can be used, alone or in combination, to implement the resources and services in different embodiments, including general purpose or special purpose computer servers, storage devices, network devices, etc. In at least some embodiments, it is a server or computing device that implements at least a portion of any one or more of the techniques described herein, including techniques for implementing the functionality of aspects discussed herein.

[0176] Aspects The following aspects are illustrative only and do not limit the scope of the disclosure or the appended claims.

[0177] Aspect 1. A computer-implemented method for detecting one or more analytes in an immunoassay, the method comprising: training a machine learning network using immunoassay reference data comprising a set of reference input parameters and corresponding reference analyte detection data; receiving a set of experimental input parameters, the experimental input parameters comprising an identifier for the analyte; applying the machine learning network to the immunoassay reference data to identify target parameters based on the analyte; and determining a set of immunoassay parameters based on the target parameters, the set of immunoassay parameters comprising a loading concentration of at least one of the analyte and a detection reagent.

[0178] Embodiment 2. The method of embodiment 1, wherein applying the machine learning network further comprises extracting relevant immunoassay reference data based on the experimental input parameters, determining a relationship between the experimental input parameters and corresponding reference analyte detection data, and predicting target parameters for detecting the analyte.

[0179] Embodiment 3. The method of embodiment 2, further comprising classifying relevant immunoassay reference data into variables, applying a statistical model to determine a relationship between two or more variables, and training a machine learning network using the relationship between the two or more variables.

[0180] Aspect 4. The method of aspect 3, wherein the statistical model comprises at least one of a cost function, a logistic regression model, a multivariate regression model, and a stochastic gradient model.

[0181] Aspect 5. The method of aspect 3, wherein the variables include one or more of antibody clonality, protein, antigen, analyte size, protein source, cell line, tissue type, detection sensitivity, loading concentration, protein abundance, antibody effect, membrane effect, blocking effect, extraction effect, protein instability, membrane type, analyte abundance, antibody binding variation, antibody clonality, binding affinity, antibody isoform specificity, backbone type, detection label, enzyme, detection multiplicity or detection sensitivity, type of precast gel (e.g. polyacrylamide gel or other gel capable of separating proteins by electrophoresis), type of membrane to be transferred, transfer method, transfer buffer, wash protocol or wash buffer.

[0182] Embodiment 6. The method of embodiment 2, further comprising determining a subclass of the analyte, identifying immunoassay reference data associated with the subclass of the analyte, determining one or more subclasses of the analyte of interest, and updating target parameters based on the detected subclass of the analyte, wherein the subclass of the analyte is a transmembrane protein, an unstable protein, a phosphorylation modification, a glycosylation modification, a post-translational modification, a protein form, a protein isoform, a truncation variant of a protein, or a mutant of a protein.

[0183] Embodiment 7. The method of embodiment 2, wherein the target parameters include at least one of a cell line, a lysate, and a gel type.

[0184] Embodiment 8. The method of any one of embodiments 1 to 7, wherein the immunoassay reference data comprises at least one of Western blot data, multiple Western blot captures, quantitative data (wherein the quantitative data optionally represents the relative abundance of proteins), categorical data, protein detection data, and immunocytometry data.

[0185] Embodiment 9. The method of embodiment 8, wherein the quantitative data comprises analyte abundance estimates.

[0186] Embodiment 10. The method of embodiment 8, wherein the immunocytometry data comprises at least one of analyte localization data or analyte intensity data.

[0187] Embodiment 11. The method of any one of embodiments 1 to 10, wherein the analyte is at least one of a hapten, a hormone, a nucleic acid, a peptide, a modified peptide, or a modified form of any of the preceding analytes.

[0188] Embodiment 12. The method according to embodiment 11, wherein the modified peptide is formed by at least one of methylation and acetylation.

[0189] Embodiment 13. The method of any one of embodiments 1-12, wherein the immunoassay parameter set comprises one or more of cell line, analyte loading range, target protein, detection data, lysate type, lysate loading concentration, protein clonality, antibody type, antibody binding site, antibody clonality, antibody dilution range, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection reagent, detection multiplicity, detection sensitivity, target range of loading concentrations for at least one of the analytes, clonality, antibody type, detection reagent, immunoassay performance prediction, recommended protein source, optimal cell line, cell source, antibody clonality, detection technique, clonality recommendation, antibody recommendation, analyte recommendation, analyte source recommendation, gel type recommended detection reagent, antibody type, antibody loading concentration, protein lysate concentration, target cell line, target protein source, analyte amount, migration conditions, validation flags, and predicted analyte locations.

[0190] Embodiment 14. The method of embodiment 13, wherein the immunoassay parameter set comprises a clonality recommendation, and the clonality recommendation comprises at least one of a monoclonal analyte recommendation or a polyclonal analyte recommendation.

[0191] Aspect 15. The method of aspect 13, wherein the detection technique applies at least one of chemiluminescence, fluorescence, enzyme, and colorimetric analysis.

[0192] Aspect 16. The method of any one of aspects 1 to 15, wherein the machine learning network includes at least one of a regression model, a decision tree based training model, and a stochastic gradient descent model.

[0193] Embodiment 17. The method of any one of embodiments 1 to 16, wherein the experimental input parameters comprise one or more of analyte type, protein type, clonality, cell line, detection reagent, antibody concentration, substrate, substrate sensitivity, detection sensitivity, lysate concentration, cell line, set of proteins, antibody binding change, blocking data, cell lysate preparation data, protein instability, protein stability, gel parameters, analyte amount range, protein mass range, cell line, detection data, lysate type, lysate loading concentration, protein, protein isoform, fragment of protein or post-translationally modified protein, antibody binding site, antibody clonality, antibody dilution, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection multiplicity or detection sensitivity.

[0194] Embodiment 18. The method of embodiment 17, wherein the detection data comprises chemical detection data (e.g., chemical substrates), biochemical detection data (e.g., enzymes), sequence-based detection, amplification-based detection, and / or fluorescence detection data.

[0195] Embodiment 19. The method of any one of embodiments 1 to 18, wherein the experimental input parameters include an analyte source, a set of proteins, and a cell line.

[0196] Aspect 20. The method of any one of aspects 1 to 19, wherein the experimental input parameters include a set of constraints received via a user interface, and the immunoassay parameter set includes recommended values ​​for the set of constraints.

[0197] Embodiment 21. The method of any one of embodiments 1 to 20, wherein the experimental input parameters include data from a plurality of assays.

[0198] Embodiment 22. The method of any one of embodiments 1 to 21, wherein the experimental input parameters include at least one detection protein or target analyte, and the immunoassay parameter set includes a loading recommendation for the at least one detection protein or target analyte.

[0199] Aspect 23. The method of any one of aspects 1 to 22, wherein the experimental input parameters include a set of proteins and a set of constraints received via a user interface, and the target parameter specifies a target value for at least one constraint in the set of constraints.

[0200] Embodiment 24. The method of embodiment 23, wherein the set of constraints comprises at least one of available lysates, cell lines, tissue types, detection techniques, and antibody clonality.

[0201] Embodiment 25. The method of any one of embodiments 1 to 24, further comprising extracting immunoassay training and reference data from images representing experimental immunoassay data.

[0202] Aspect 26. The method of any one of aspects 1 to 25, further comprising receiving information indicating an immunoassay parameter set type, the immunoassay parameter set type being at least one of a troubleshooting solution or an experimental design, and updating the target parameters based on the immunoassay parameter set type.

[0203] Embodiment 27. The method according to embodiment 26, wherein the target parameter relates to at least one of a Western blot application, an immunoblotting method, a transfer method, and a type of protein gel.

[0204] Aspect 28. The method of aspect 26, wherein the troubleshooting solution identifies one or more of the analyte source, antibody, and dilution detection information.

[0205] Aspect 29. A system for detecting one or more analytes in an immunoassay, comprising at least one computing device comprising a processor and at least one memory storing instructions, which, when executed by the processor, cause the computing device to receive immunoassay reference data comprising a set of input parameters and corresponding analyte detection data, receive a set of experimental input parameters, the experimental input parameters comprising an analyte identifier and clonality, apply a machine learning network to the immunoassay reference data to identify target parameters based on the analyte, determine an immunoassay parameter set based on the target parameters, and provide an immunoassay parameter set, the immunoassay parameter set comprising a loading concentration of at least one of the analyte and the detection reagent.

[0206] Aspect 30. The system of aspect 29, wherein the at least one memory, when executed by the processor, further causes the computing device to extract relevant immunoassay reference data based on the experimental input parameters, determine a relationship between the experimental input parameters and corresponding reference analyte detection data, and predict target parameters for detecting the analyte.

[0207] Aspect 31. The system of aspect 30, wherein the at least one memory stores instructions that, when executed by the processor, further cause the computing device to classify associated immunoassay reference data into variables, apply a statistical model to determine a relationship between two or more variables, and train a machine learning network using the relationship between the two or more variables.

[0208] Embodiment 32. The system of any one of embodiments 29 to 31, wherein the immunoassay reference data comprises analyte detection data from a plurality of experiments comprising one or more of different loading concentrations, different detection reagents, and different antibody affinities.

[0209] Aspect 33. A system according to any one of aspects 29 to 32, wherein the immunoassay parameter set comprises a target range of loading concentrations of at least one of the analyte and the detection reagent.

[0210] Aspect 34. The system of any one of aspects 29 to 33, wherein the immunoassay parameter set further includes at least one of an immunoassay performance prediction, a recommended analyte, a recommended detection reagent, an antibody concentration, a protein lysate concentration, an antibody type, a target cell line, a target protein source, an analyte amount, a transfer condition, a validation flag, and a predicted analyte location.

[0211] Aspect 35. The system of any one of aspects 29 to 34, wherein the machine learning network applies at least one of a regression model, a decision tree, and a stochastic gradient descent model.

[0212] Aspect 36. The system of any one of aspects 29 to 35, wherein the experimental input parameters include one or more of protein, clonality, cell line, detection agent, antibody concentration, substrate, substrate sensitivity, detection sensitivity, lysate concentration, cell line, set of proteins, and antibody binding variance.

[0213] Embodiment 37. The system of any one of embodiments 29 to 36, wherein the instructions further cause the computing device to extract immunoassay reference data from an image representing the experimental immunoassay data.

[0214] Aspect 38. The system of any one of aspects 29 to 37, further comprising a user interface, the user interface comprising at least one of an equipment console, a web tool, a graphical user interface, and a display on a computing device.

[0215] Aspect 39. A non-transitory computer-readable medium for storing instructions that, when executed by one or more processors, cause a device to perform a method according to any one of aspects 1-30.

[0216] A device comprising: one or more processors; and a memory storing instructions that, when executed by the one or more processors, cause the device to perform a method according to any one of the embodiments 1-39.

[0217] Aspect 41. A computer-implemented method for operating an immunoassay instrument support device, the method comprising: receiving immunoassay reference data including a set of reference input parameters and corresponding reference analyte detection data; receiving a set of experimental input parameters, the experimental input parameters including at least one variable associated with an immunoassay experiment; receiving information indicative of a problem related to at least one of the immunoassay reference data, the experimental input parameters, and a result of the immunoassay experiment; applying a machine learning network to the immunoassay reference data to identify target parameters based on the problem; and determining a solved immunoassay parameter set based on the target parameters.

[0218] Aspect 42. The method of aspect 41, wherein the problem is at least one of an analyte detection problem or an immunoassay reference data extraction problem.

[0219] Aspect 43. The method of aspect 42, wherein the problem concerns the extraction of immunoassay reference data from an image showing experimental immunoassay data.

[0220] Embodiment 44. The method of any one of embodiments 41 to 43, wherein the immunoassay parameter set provides at least one of protein lysate from a cell line source, type of lysate, antibody dilution range for target detection, antibody dilution based on clonality and host backbone type, type of detection reagent, type of detection technology, and target detection technology to ensure linearity.

[0221] Embodiment 45. The method of any one of embodiments 41 to 44, wherein the experimental input parameters include at least one of analyte type, protein type, clonality, cell line, detection reagent, antibody concentration, substrate, substrate sensitivity, detection sensitivity, lysate concentration, cell line, set of proteins, antibody binding variance, blocking data, cell lysate preparation data, protein instability, protein stability, gel parameters, analyte amount range, and protein mass range.

[0222] Embodiment 46. The method of any one of embodiments 41 to 45, wherein the experimental input parameters include one or more of a hapten, a hormone, a modified nucleic acid, a peptide, an antibody clonality, a protein, an antigen, an analyte size, a protein source, a cell line, a tissue type, a detection sensitivity, a loading concentration, a protein abundance, an antibody effect, a membrane effect, a blocking effect, an extraction effect, a protein instability, a membrane type, an analyte abundance, an antibody binding change, an antibody clonality, a binding affinity, an antibody isoform specificity, a backbone type, a detection label, an enzyme, a detection multiplicity, or a detection sensitivity.

[0223] Aspect 47. A method according to any one of aspects 41 to 46, wherein the experimental input parameters include a set of variables received via a user interface, the problem relates to detection, and the immunoassay parameter set includes recommended values ​​for the set of variables.

[0224] Embodiment 48. The method of any one of embodiments 41 to 47, wherein the immunoassay parameter set comprises one or more of cell line, analyte loading range, target protein, detection data, lysate type, lysate loading concentration, antibody type, antibody binding site, antibody clonality, antibody dilution range, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection reagent, detection multiplicity or detection sensitivity.

[0225] Embodiment 49. The method of any one of embodiments 41 to 48, wherein the set of immunoassay parameters includes one or more of a target range of loading concentration of at least one of the analytes, an antibody type, a detection reagent, an immunoassay performance prediction, a recommended protein source, an optimal cell line, a cell source, an antibody clonality, a detection technique, a clonality recommendation, an antibody recommendation, an analyte recommendation, an analyte source recommendation, a recommended detection reagent, an antibody type, an antibody loading concentration, a protein lysate concentration, a target protein source, an analyte amount, a transfer condition, a validation flag, and a predicted analyte location.

[0226] Aspect 50. The method of any one of aspects 41 to 49, wherein the machine learning network applies at least one of a regression model, a decision tree-based training model, and a stochastic gradient descent model.

[0227] Aspect 51. A computer-implemented method for operating an immunoassay instrument support device, the method comprising: receiving immunoassay reference data including a set of reference input parameters and corresponding reference analyte detection data; receiving a set of experimental input parameters; receiving at least one target variable for an immunoassay experiment; applying a machine learning network to the immunoassay reference data to identify a target parameter based on the target variables; and determining a set of immunoassay parameters for obtaining the target variable based on the target parameters, wherein the target variable is an analyte.

[0228] Aspect 52. The method of aspect 51, wherein the set of experimental input parameters represents a method of the experiment.

[0229] Aspect 53. The method according to aspect 51 or 52, wherein the experimental method is a Western blot method, an immunoblotting method, a transfer method, or a method using a protein gel.

[0230] Embodiment 54. The method of any one of embodiments 51 to 53, wherein the at least one target variable comprises one or more of analyte type, protein type, clonality, cell line, detection reagent, antibody concentration, substrate, substrate sensitivity, detection sensitivity, lysate concentration, cell line, set of proteins, antibody binding change, blocking data, cell lysate preparation data, protein instability, protein stability, gel parameters, analyte amount range, and protein mass range.

[0231] Embodiment 55. The method of any one of embodiments 51 to 54, wherein the at least one target variable comprises one or more of a hapten, a hormone, a modified nucleic acid, a peptide, a protein, an antigen, an analyte size, a protein source, a cell line, a tissue type, detection sensitivity, a loading concentration, a protein abundance, an antibody effect, a membrane effect, a blocking effect, an extraction effect, a protein instability, a membrane type, an analyte abundance, an antibody binding change, an antibody clonality, a binding affinity, an antibody isoform specificity, a backbone type, a detection label, an enzyme, a detection multiplicity or a detection sensitivity.

[0232] Embodiment 56. The method of any one of embodiments 51 to 55, wherein the immunoassay parameter set comprises one or more of cell line, analyte loading range, target protein, detection data, lysate type, lysate loading concentration, antibody type, antibody binding site, antibody clonality, antibody dilution range, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection reagent, detection multiplicity or detection sensitivity.

[0233] Embodiment 57. The method of any one of embodiments 51 to 56, wherein the immunoassay parameter set includes one or more of a target range of loading concentration of at least one of the analytes, an antibody type, a detection reagent, an immunoassay performance prediction, a recommended protein source, an optimal cell line, a cell source, an antibody clonality, a detection technique, a clonality recommendation, an antibody recommendation, an analyte recommendation, an analyte source recommendation, a recommended detection reagent, an antibody type, an antibody loading concentration, a protein lysate concentration, a target protein source, an analyte amount, a transfer condition, a validation flag, and a predicted analyte location.

[0234] Aspect 58. The method of any one of aspects 51 to 57, wherein the machine learning network applies at least one of a regression model, a decision tree-based training model, and a stochastic gradient descent model.

[0235] Aspect 59. A computer-implemented method for operating an immunoblot instrument support apparatus, the method comprising: receiving immunoblot reference data comprising a set of reference input parameters and corresponding reference analyte detection data; receiving a set of experimental input parameters, the experimental input parameters comprising an analyte; applying a machine learning network to the immunoblot reference data to identify target parameters based on the analyte; and determining a set of immunoassay parameters based on the target parameters, the set of immunoassay parameters comprising a loading concentration of at least one of the analyte and a detection reagent.

[0236] Aspect 60. The method of aspect 59, wherein the immunoassay parameter set comprises one or more of cell line, analyte loading range, target protein, detection data, lysate type, lysate loading concentration, antibody type, antibody binding site, antibody clonality, antibody dilution range, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection reagent, detection multiplicity, detection sensitivity, target range of loading concentration of at least one of the analytes, clonality, antibody type, detection reagent, immunoassay performance prediction, recommended protein source, optimal cell line, cell source, antibody clonality, detection technique, clonality recommendation, antibody recommendation, analyte recommendation, analyte source recommendation, gel type recommended detection reagent, antibody type, antibody loading concentration, protein lysate concentration, target protein source, analyte amount, migration conditions, validation flags, and predicted analyte location.

[0237] Embodiment 61. The method of embodiment 59 or 60, wherein the experimental input parameters comprise one or more of analyte type, protein type, cell line, detection reagent, antibody concentration, substrate, substrate sensitivity, detection sensitivity, lysate concentration, cell line, set of proteins, antibody binding change, blocking data, cell lysate preparation data, protein instability, protein stability, gel parameters, analyte amount range, protein mass range, cell line, detection data, lysate type, lysate loading concentration, protein, protein isoform, fragment of protein or post-translationally modified protein, antibody binding site, antibody clonality, antibody dilution, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection multiplicity or detection sensitivity.

[0238] Aspect 62. A computer-implemented method for operating an immunoassay instrument support device, the method comprising: receiving immunoassay reference data comprising a set of reference input parameters and corresponding reference analyte detection data; receiving a set of experimental input parameters, the experimental input parameters comprising an analyte; applying a machine learning network to the immunoassay reference data to identify target parameters comprising a set of multiple proteins for profiling based on the analyte; and determining a set of immunoassay parameters for profiling the set of multiple proteins based on the target parameters, the set of immunoassay parameters comprising a loading concentration of at least one of the analyte and a detection reagent.

[0239] Embodiment 63. The computer-implemented method of embodiment 62, wherein the immunoassay is a bead-based immunoassay.

[0240] Embodiment 64. The computer-implemented method of embodiment 63, wherein the bead-based immunoassay is at least one of a multiplex assay, a panel-based bead-based immunoassay, or an activated surface panel builder-based bead-based immunoassay.

[0241] Embodiment 65. The method of any one of embodiments 62-64, wherein the set of immunoassay parameters comprises one or more of cell line, analyte loading range, target protein, detection data, lysate type, lysate loading concentration, antibody type, antibody binding site, antibody clonality, antibody dilution range, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection reagent, detection multiplicity, detection sensitivity, target range of loading concentration of at least one of the analytes, clonality, antibody type, detection reagent, immunoassay performance prediction, recommended protein source, cell source, antibody clonality, detection technique, clonality recommendation, antibody recommendation, analyte recommendation, analyte source recommendation, gel type recommended detection reagent, antibody type, antibody loading concentration, protein lysate concentration, target cell line, target protein source, analyte amount, migration conditions, validation flags, and predicted analyte location.

[0242] Embodiment 66. The method of any one of embodiments 62 to 65, wherein the experimental input parameters include one or more of analyte type, protein type, cell line, detection reagent, antibody concentration, substrate, substrate sensitivity, detection sensitivity, lysate concentration, cell line, set of proteins, antibody binding change, blocking data, cell lysate preparation data, protein instability, protein stability, gel parameters, analyte amount range, protein mass range, cell line, detection data, lysate type, lysate loading concentration, protein, protein isoform, fragment of protein or post-translationally modified protein, antibody binding site, antibody clonality, antibody dilution, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection multiplicity, or detection sensitivity.

[0243] Aspect 67. A computer-implemented method for operating an immunoassay instrument support device, the method comprising: receiving immunoassay reference data including a set of reference input parameters and corresponding reference analyte detection data; receiving a set of experimental input parameters, the experimental input parameters including analyte and single cell information (e.g., information related to an individual or a single cell); applying a machine learning network to the immunoassay reference data to identify target parameters based on the analyte; and determining a set of immunoassay parameters for detecting multiple proteins in a single cell based on the target parameters, the immunoassay parameter set including a loading concentration of at least one of the analyte and the detection reagent.

[0244] Aspect 68. The method of aspect 67, wherein the immunoassay parameter set comprises one or more of cell line, analyte loading range, target protein, detection data, lysate type, lysate loading concentration, antibody type, antibody binding site, antibody clonality, antibody dilution range, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection reagent, detection multiplicity, detection sensitivity, target range of loading concentration of at least one of the analytes, clonality, antibody type, detection reagent, immunoassay performance prediction, recommended protein source, optimal cell line, cell source, antibody clonality, detection technique, clonality recommendation, antibody recommendation, analyte recommendation, analyte source recommendation, gel type recommended detection reagent, antibody type, antibody loading concentration, protein lysate concentration, target protein source, analyte amount, migration conditions, validation flag, and predicted analyte location.

[0245] Embodiment 69. The method of embodiment 67 or 68, wherein the experimental input parameters comprise one or more of analyte type, protein type, cell line, detection reagent, antibody concentration, substrate, substrate sensitivity, detection sensitivity, lysate concentration, cell line, set of proteins, antibody binding change, blocking data, cell lysate preparation data, protein instability, protein stability, gel parameters, analyte amount range, protein mass range, cell line, detection data, lysate type, lysate loading concentration, protein, protein isoform, fragment of protein or post-translationally modified protein, antibody binding site, antibody clonality, antibody dilution, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection multiplicity or detection sensitivity.

[0246] Aspect 70. A computer-implemented method for operating an immunoassay instrument support apparatus for a flow-based immunoassay, comprising: receiving immunoassay reference data including a set of reference input parameters and corresponding reference analyte detection data, the immunoassay reference data including at least one set of flow-based immunoassay data; receiving a set of experimental input parameters, the experimental input parameters including analyte and single cell information; applying a machine learning network to the immunoassay reference data including the flow-based immunoassay data to identify target parameters based on the analyte; and determining an immunoassay parameter set for detecting multiple proteins in a single cell based on the target parameters, the immunoassay parameter set including a loading concentration of at least one of the analyte and the detection reagent. Examples of flow assays include, for example, flow assays using colored particles, competitive assays, etc.

[0247] Embodiment 71. The method according to embodiment 70, wherein the flow-based immunoassay is a lateral flow immunoassay.

[0248] Aspect 72. The method of aspect 70 or 71, wherein the set of immunoassay parameters comprises one or more of cell line, analyte loading range, target protein, detection data, lysate type, lysate loading concentration, antibody type, antibody binding site, antibody clonality, antibody dilution range, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection reagent, detection multiplicity, detection sensitivity, target range of loading concentration of at least one of the analytes, clonality, antibody type, detection reagent, immunoassay performance prediction, recommended protein source, optimal cell line, cell source, antibody clonality, detection technique, clonality recommendation, antibody recommendation, analyte recommendation, analyte source recommendation, gel type recommended detection reagent, antibody type, antibody loading concentration, protein lysate concentration, target protein source, analyte amount, migration conditions, validation flag, and predicted analyte location.

[0249] Embodiment 73. The method of any one of embodiments 70 to 72, wherein the experimental input parameters comprise one or more of analyte type, protein type, cell line, detection reagent, antibody concentration, substrate, substrate sensitivity, detection sensitivity, lysate concentration, cell line, set of proteins, antibody binding change, blocking data, cell lysate preparation data, protein instability, protein stability, gel parameters, analyte amount range, protein mass range, cell line, detection data, lysate type, lysate loading concentration, protein, protein isoform, fragment of protein or post-translationally modified protein, antibody binding site, antibody clonality, antibody dilution, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection multiplicity or detection sensitivity.

[0250] Aspect 74. A computer-implemented method for optimizing fluorescent analyte detection in an immunoassay, the method comprising: receiving reference data comprising a set of reference input parameters and corresponding reference analyte detection data; receiving a set of experimental input parameters, the experimental input parameters comprising an analyte; applying a machine learning network to the reference data to identify target parameters for fluorescent detection based on the analyte; and determining an immunoassay parameter set based on the target parameters, the immunoassay parameter set comprising a loading concentration of at least one of the analyte and a detection reagent.

[0251] Embodiment 75. The method of embodiment 74, wherein the immunoassay parameter set further comprises a type of dye for detecting at least one of the protein or analyte.

[0252] Embodiment 76. The method of embodiment 75, wherein the type of dye is at least one of an Ab conjugate, a secondary Ab conjugate, or a strong dye for detecting low amounts of an analyte.

[0253] Embodiment 77. The method of any one of embodiments 74-76, wherein the set of immunoassay parameters comprises one or more of cell line, analyte loading range, target protein, detection data, lysate type, lysate loading concentration, antibody type, antibody binding site, antibody clonality, antibody dilution range, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection reagent, detection multiplicity, detection sensitivity, target range of loading concentration of at least one of the analytes, clonality, antibody type, detection reagent, immunoassay performance prediction, recommended protein source, optimal cell line, cell source, antibody clonality, detection technique, clonality recommendation, antibody recommendation, analyte recommendation, analyte source recommendation, gel type recommended detection reagent, antibody type, antibody loading concentration, protein lysate concentration, target protein source, analyte amount, migration conditions, validation flags, and predicted analyte location.

[0254] Embodiment 78. The method of any one of embodiments 74 to 77, wherein the experimental input parameters comprise one or more of analyte type, protein type, cell line, detection reagent, antibody concentration, substrate, substrate sensitivity, detection sensitivity, lysate concentration, cell line, set of proteins, antibody binding change, blocking data, cell lysate preparation data, protein instability, protein stability, gel parameters, analyte amount range, protein mass range, cell line, detection data, lysate type, lysate loading concentration, protein, protein isoform, fragment of protein or post-translationally modified protein, antibody binding site, antibody clonality, antibody dilution, binding affinity, antibody isoform specificity, backbone type, protein stability, protein instability, detection label, enzyme, detection multiplicity or detection sensitivity.

[0255] Aspect 79. A method comprising: receiving experimental data corresponding to an immunoassay experiment, the experimental data being based on an analyte of interest; analyzing the experimental data by a machine learning process trained to determine the extent to which an analyte band shifts in an immunoassay; and determining the extent of band shift in the immunoassay experiment based on the analysis, the band comprising the analyte of interest.

[0256] Aspect 80. The method of aspect 79, wherein the machine learning process includes a neural network.

[0257] Aspect 81. The method of aspect 80, wherein the neural network is a feedforward network or a deep neural network.

[0258] Embodiment 82. The method of embodiment 79, wherein analyzing includes analyzing an immunoassay image of the immunoassay experiment, and determining includes marking the degree of shift of the band on the immunoassay image.

[0259] Aspect 83. The method of aspect 82, wherein the immunoassay image comprises a stained gel or a capillary gel loaded with the analyte.

[0260] Embodiment 84. The method according to embodiment 79, wherein the immunoassay experiment is a bead-based immunoassay or a flow-based immunoassay.

[0261] Embodiment 85. The method according to embodiment 79, wherein the analyte is at least one of a protein, a hapten, a hormone, a nucleic acid, a peptide, a modified peptide, or a modified form of any of the preceding analytes.

[0262] Aspect 86. The method of aspect 79, wherein the machine learning process has been trained using an immunoassay dataset that includes at least one of glycosylation, disulfide bond, modified residue, 3-(N-morpholino)propanesulfonic acid (MOPS), ubiquitination, lipidation, 2-(N-morpholino)ethanesulfonic acid (MES), isoelectric point (pI), sumoylation, Tris acetate, interchain or polymer crosslinks, gel type, buffer type, or degree of shift.

[0263] Embodiment 87. The method of embodiment 79 or embodiment 86, wherein the experimental data comprises at least one of glycosylation, disulfide bonds, modified residues, 3-(N-morpholino)propanesulfonic acid (MOPS), ubiquitination, lipidation, 2-(N-morpholino)ethanesulfonic acid (MES), isoelectric point (pI), sumoylation, Tris acetate, or interchain or polymer crosslinks of the analyte of interest, as well as gel type and buffer type for immunoassay experiments.

[0264] Embodiment 88. The method according to embodiment 79, wherein the degree of band shift is caused by a shift in molecular weight corresponding to the analyte of interest.

[0265] Aspect 89. The method of aspect 79, further comprising determining whether the experimental data should be modified and displaying the determined modification.

[0266] Aspect 90. A system comprising a processor and a memory coupled to the processor, the memory storing instructions that, when executed by the processor, cause the processor to perform operations including receiving experimental data, the experimental data including an analyte identifier and an immunoassay image; analyzing the immunoassay image to mark features of the immunoassay image, the features including bands; determining by a machine learning process an extent of band shift of a band on the immunoassay image, the band including an analyte of interest corresponding to the identifier; and displaying the extent of band shift on the immunoassay image.

[0267] Aspect 91. The system described in aspect 90, wherein the machine learning process includes a neural network.

[0268] Aspect 92. The system described in aspect 91, wherein the neural network is a feedforward network or a deep neural network.

[0269] Aspect 93. The system of aspect 90, wherein the processor performs further operations of calculating molecular weight data for the bands and including a data table, the data table including the calculated molecular weights.

[0270] Aspect 94. The system of aspect 90, wherein the experimental data further comprises at least one identifier of the cell line, an identifier of the molecular marker, a type of lysate, a loading concentration, and a type of gel.

[0271] Aspect 95. The system of aspect 90, wherein the processor performs further operations including determining whether the experimental data should be modified and displaying the determined modification.

[0272] Aspect 96. The system described in aspect 90, wherein the immunoassay image is an image of a bead-based immunoassay or a flow-based immunoassay.

[0273] Aspect 97. The system according to aspect 90, wherein the analyte is at least one of a protein, a hapten, a hormone, a nucleic acid, a peptide, a modified peptide, or a modified form of any of the preceding analytes.

[0274] Aspect 98. The system described in aspect 90, wherein the feature further includes at least one of a frame of an immunoassay image and a lane of an immunoassay image.

[0275] Aspect 99. The system of aspect 90, wherein the immunoassay image comprises a plurality of bands.

[0276] Embodiment 100. The system of embodiment 98, wherein the immunoassay image comprises a plurality of lanes.

[0277] Embodiment 101. The system of embodiment 99, wherein each lane in the plurality of lanes is analyzed to determine a band shift in each lane.

[0278] Aspect 102. The system described in aspect 90, wherein the processor performs an operation further including displaying whether a band was found, was found along with a non-specific band, was not found, or no band was present.

[0279] Aspect 103. The system of aspect 90, wherein the machine learning process is trained using an immunoassay dataset to determine the degree to which an analyte band shifts in an immunoassay experiment.

[0280] Aspect 104. The system of aspect 103, wherein the immunoassay dataset comprises at least one of glycosylation, disulfide bonds, modified residues, 3-(N-morpholino)propanesulfonic acid (MOPS), ubiquitination, lipidation, 2-(N-morpholino)ethanesulfonic acid (MES), isoelectric point (pI), SUMOylation, Tris acetate, interchain or polymer crosslinks, gel type, and buffer type.

[0281] Aspect 105. The system described in aspect 90, wherein the receiving and displaying are performed by a user interface.

Claims

1. Obtaining experimental data corresponding to an immunoassay experiment, the experimental data being based on an analyte of interest; analyzing the experimental data with a machine learning process trained to determine the extent to which the analyte band shifts in an immunoassay; determining the degree of shift for a band in the immunoassay experiment based on the analyzing, the band containing the analyte of interest; A method comprising:

2. The method of claim 1 , wherein the machine learning process comprises a neural network, the neural network comprising a feedforward network or a deep neural network.

3. said analyzing comprising analyzing immunoassay images of said immunoassay experiments; said determining includes marking the degree of shift for said band on said immunoassay image; The method of claim 1 , wherein the immunoassay image comprises a stained gel or a capillary gel into which the analyte is loaded.

4. The method of claim 1 , wherein the immunoassay experiment is a bead-based immunoassay or a flow-based immunoassay.

5. 10. The method of claim 1, wherein the analyte is at least one of a protein, a hapten, a hormone, a nucleic acid, a peptide, a modified peptide, or a modified form of any of the foregoing analytes.

6. 2. The method of claim 1, wherein the machine learning process is trained using an immunoassay dataset that includes at least one of glycosylation, disulfide bonds, modified residues, 3-(N-morpholino)propanesulfonic acid (MOPS), ubiquitination, lipidation, 2-(N-morpholino)ethanesulfonic acid (MES), isoelectric point (pI), sumoylation, Tris acetate, interchain or polymer crosslinks, gel type, buffer type, or degree of shift.

7. 10. The method of claim 1, wherein the experimental data includes at least one of glycosylation, disulfide bonds, modified residues, 3-(N-morpholino)propanesulfonic acid (MOPS), ubiquitination, lipidation, 2-(N-morpholino)ethanesulfonic acid (MES), isoelectric point (pI), sumoylation, Tris acetate, or interchain or polymer crosslinks of the analyte of interest, as well as gel type and buffer type for the immunoassay experiment.

8. 2. The method of claim 1, wherein the degree of shift for the band is caused by a shift in molecular weight corresponding to the analyte of interest.

9. determining whether the experimental data should be corrected; The method of claim 1 , further comprising: displaying the determined modification.

10. 1. A system comprising: at least one processor; a memory coupled to the at least one processor, the memory, when executed by the processor, causing the processor to: acquiring experimental data including analyte identifiers and immunoassay images; analyzing the immunoassay image to mark features of the immunoassay image including at least one band; determining, by a machine learning process, the degree of shift for a band on the immunoassay image, the band containing the analyte of interest corresponding to the identifier; The system stores instructions for performing operations including displaying the degree of shift for the band on the immunoassay image.

11. The system of claim 10 , wherein the machine learning process comprises a neural network.

12. The processor: Calculating the molecular weight data of the bands; The system of claim 10 , further performing an action including displaying a data table containing the calculated molecular weights.

13. The system of claim 10 , wherein the experimental data further comprises at least one identifier of a cell line, an identifier of a molecular marker, a lysate type, a loading concentration, and a gel type.

14. The processor: determining whether the experimental data should be corrected; The system of claim 10 , further performing an action including displaying the determined modification.

15. The system of claim 10 , wherein the immunoassay image is an image of a bead-based immunoassay or a flow-based immunoassay.

16. The system of claim 10 , wherein the analyte is at least one of a protein, a hapten, a hormone, a nucleic acid, a peptide, a modified peptide, or a modified form of any of the foregoing analytes.

17. 11. The system of claim 10, wherein the features further include at least one of a frame of the immunoassay image and at least one lane of the immunoassay image, the at least one lane being analyzed to determine the shift of the band in each lane.

18. The at least one processor 11. The system of claim 10, further performing an operation including displaying whether the band was found, found with a non-specific band, not found, or no band.

19. 11. The system of claim 10, wherein the machine learning process is trained using an immunoassay dataset to determine the degree to which an analyte band shifts in an immunoassay experiment.

20. 20. The system of claim 19, wherein the immunoassay dataset comprises at least one of glycosylation, disulfide bond, modified residue, 3-(N-morpholino)propanesulfonic acid (MOPS), ubiquitination, lipidation, 2-(N-morpholino)ethanesulfonic acid (MES), isoelectric point (pI), sumoylation, Tris acetate, interchain or polymer crosslinks, gel type, and buffer type.

Images