Abstract: Gene expression vectors of two samples in a gene expression space can be transformed into gene feature vectors in a gene feature space. Gaussian densities and associated weights can be determined for each sample using the corresponding gene feature vectors. Gaussian densities of one samples can be aligned to Gaussian densities of the other sample. A difference between a pair of aligned Gaussian densities of the two samples can be determined, which indicates a difference between subpopulation of cells of the two samples represented by the Gaussian densities of the pair of aligned Gaussian densities.

Abstract: The present disclosure provides methods and systems of identifying an inflammatory disease or condition, e.g., an inflammatory bowel disease in a subject using a DeepLearning model. The DeepLearning model may be used to predict, treat, monitor, and/or prevent the inflammatory disease or condition in the subject, as well as to characterize a subtype of the inflammatory disease or condition.

Abstract: A method is described for real-time detecting and classifying of a characteristic signal, such as a neural spike, and forwarding information for further processing if it meets certain criteria. A system (e.g., an on-chip system implanted in a subject's cranium) receives an electrical biological signal. The system filters the signal to generate a filtered signal and fits the filtered signal to a model. The system identifies a set of fit values based on the model, the set of fit values comprising a plurality of sample amplitude values and a respective plurality of time values. Based on the fit values, the system computes a set of characteristic values. The system compares the characteristic values to a corresponding set of threshold values. Based on the comparison, the system determines whether the received biological signal corresponds to a neural spike and, if a spike is detected, forwards on information.

Abstract: The present disclosure provides methods for non-invasive prenatal screening (NIPS) of fetal aneuploidies. The present methods are based on analyzing cell-free fetal DNA (cff DNA) found in a pregnant woman's circulation through the next generation sequencing (NGS) technology. Particularly, the present methods analyze the relative abundance of different fetal genomic fragments present in the maternal sample, where the fragments can be aligned to particular chromosomal locations of the fetal genome. The relative abundance information is indicative as to whether a particular chromosome is overrepresented or underrepresented in a fetal genome as compared to normal individuals, and thus can be used to detect fetal aneuploidy. Additionally, methods for increasing the positive predictive values (PPV) of NIPS by excluding false-positive detections are also provided.

Abstract: Technologies are described related to determining protein structure and properties based on sequences of proteins. In various implementations, a first model can be generated to determine structural features of proteins based on amino acid sequences of the proteins. Additionally, a second model can be generated to determine biophysical properties of proteins based on structural features of the proteins. In particular implementations, an amino acid sequence of a particular protein can be utilized by the first model to determine one or more structural features of the protein. The one or more structural features of the protein generated by the first model can be utilized by the second model to determine at least one biophysical property of the protein.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods that a implement a framework for active learning to discover pairwise interactions via representation learning. Indeed, in one or more implementations, the disclosed systems generate a first individual perturbation embedding from a first representation of a first cell exposed to a first perturbation and a second individual perturbation embedding, from a second representation of a second cell exposed to a second perturbation. For instance, the disclosed systems combine the first individual perturbation embedding and the second individual perturbation embedding to determine a predicted pairwise embedding. Moreover, in some instances, the disclosed systems generate a pairwise embedding from a representation of a cell exposed to both the first and second perturbation.

Abstract: The invention provides a method of diagnosing overactive bladder disorder (OAB), the method comprising: measuring the concentrations of one or more of adenosine triphosphate (ATP), acetylcholine (ACh), nitric oxide (NO) and interleukin 5 (IL-5) in a sample obtained from a subject; normalising the concentrations to the concentration of creatinine (Cr) in the sample; range standardising the normalised concentrations and subject's age to the following values: Age to 120 years old; ATP/Cr to 0.000001; ACh/Cr to 0.1; NO to 20000; IL-5/Cr to 100; wherein the likelihood of having OAB (pOAB)=1/1+e?x, where X=one or more of the following: (a) (?2.688±1.050)+5.472±2.098×subject's age+1.356±0.559×Gender (Female=1, Male=0)+(?7.998±40.273)×[IL-5/Cr]; (b) (?2.141±0.966)+4.506±1.902×subject's age+1.034±0.519×Gender (Female=1, Male=0)+(?5294.063±9075.456)×[ACh/Cr]; (c) (?2.825±1.072)+5.964±2.167×subject's age+1.312±0.562×Gender (Female=1, Male=0)+17.790±58.762×[IL-5/Cr]+(?9180.821±12700.057)×[ACh/Cr]; (d) (?2.993±1.197)+5.

Abstract: A predictive system may access a set of physical structures corresponding to a physical system. Each physical structure representative of a configuration. The predictive system may encode the accessed physical structures to produce a set of encoded physical structures by encoding, for each accessed physical structure, a position of each constituent unit of the physical system within the accessed physical structure. The predictive system may train a machine-learned model using the encoded physical structures. The predictive system may retrain the machine-learned model by iteratively: accessing a set of two or more candidate physical structures, determining a first energy difference among the set of candidate physical structures, obtaining a second energy difference between a set of physical structures corresponding to the set of candidate physical structures using a method to calculate reference energy values.

Abstract: A system, method and computer program product for analyzing data of high dimensionality (e.g., sequence reads of nucleic acid samples in connection with a disease condition) are provided.

Abstract: A program for operating a cell culture support apparatus causes a computer to acquire a learned model, derived by performing machine learning on the basis of a set of time-series data for learning indicating a time transition of an amount of each of plural types of components constituting a medium used for cell culture and good/bad data indicating good or bad of a result of the cell culture in correspondence with the time-series data for learning, indicating a guideline of the amount, acquire time-series data for analysis indicating the time transition of the amount, derive quantitative guideline information of the amount for obtaining a good result in the cell culture, with respect to at least one of the plural types of components, from the learned model and input data of at least a part of the time-series data for analysis acquired, and output guideline information.

Abstract: The disclosure is related to methods of cancer early detection. In some embodiments, a panel of selected protein tumor markers are used for effective and affordable multi-cancer early detection.

Abstract: A machine-learning based, interactive chronic pain and food trigger tracking system and processes and a cross-platform chronic pain and food trigger tracking app are disclosed. A machine learning algorithm automatically categorizes pain symptoms against food input and categorizes the input in taxonomic categories to give users feedback on whether a food could be contributing to their pain intensity. Dialectic behavior therapy guides users through chronic pain management. The machine-learning based, interactive chronic pain and food trigger tracking system also determines possibilities of supplements helping, hurting, or not having an effect on pain intensity.

Abstract: A method, computer system, and a computer program product for designing one or more folded structural proteins from at least one raw amino acid sequence is provided. The present invention may include computing one or more character embeddings based on the at least one raw amino acid sequence by utilizing a multi-scale neighborhood-based neural network (MNNN) model. The present invention may then include refining the computed one or more character embeddings with at least one set of sequence neighborhood information. The present invention may further include predicting one or more dihedral angles based on the refined one or more character embeddings.

Abstract: Methods for producing a filtered sequencing data set by identifying one or more split sequence reads in a set of test sequence reads obtained from cell-free nucleic acid (cfNA) in a biological sample, wherein each split sequence read comprises at least one breakpoint; and, suppressing, in the set of test sequence reads, (i) at least a portion of one or more of the split sequence reads and/or at least a portion of one or more of the test sequence reads that comprise at least one sequence variant within a selected number of nucleotides from a given breakpoint, (ii) one or more base calls of the split sequence reads and/or one or more base calls of the test sequence reads that comprise at least one sequence variant within a selected number of nucleotides from a given breakpoint, thereby producing the filtered sequence information data set.

Abstract: Methods and systems for a deep-learning platform for sorting cell populations. An example method includes executing a software-platform associated with analyzing received flow cytometry data obtained via an acquisition device in communication with the computing system, and the software-platform sorting cell populations indicated in the flow cytometry data. User input is received indicating selection of a deep-learning module, the deep-learning module being obtained via a network to supplement the software-platform. The flow cytometry data is analyzed and a machine learning model is selected which was trained based on similar phenotype information as indicated in the flow cytometry data. The machine learning model is applied based on the flow cytometry data, the information being normalized based on the UMI counts associated with the flow cytometry data.

Abstract: Methods and devices involving using a bolus calculator to account for an uncertainty of a carbohydrate estimate to reduce a risk of hypoglycemia and/or a risk of hyperglycemia. More specifically, methods and devices using an algorithm executed by a processor of the bolus calculator and using a carbohydrate estimate and a measure defining the uncertainty of the carbohydrate estimate to determine the probability of hypoglycemia and/or hyperglycemia if the carbohydrate estimate is an overestimate or an underestimate, and subsequently providing recommendations and alerts to a user.

Abstract: The present invention relates to method for storing and transmitting information by employing a nucleic acid construct. The nucleic acid construct can include a lock region; a translation key region that corresponds to the identity of a key; and a message region including a nucleic sequence that corresponds to an encrypted message.