Component Gaussian Mixture Model Research Articles

An analysis of likely germline events by tumor tissue testing on large somatic panels.

e22524 Background: The increase in large panel testing has facilitated the detection of numerous events, including those in genes associated with inherited cancer. Further, large panel sequencing may uncover variants in these genes in non-inherited cancers. Hence the frequency of pathogenic/likely pathogenic (P/LP) events in some common germline genes was studied across cancers in the Indian population. Methods: A statistical model was used to derive a method to label variants as likely germline. Variants in hereditary breast and ovarian cancer (HBOC) and mismatch repair (MMR) genes from 1553 cases sequenced at our referral laboratory on the TruSight Oncology 500 panel (TSO500) were analyzed for their variant allele frequency (VAF) distribution. A 3 component Gaussian mixture model (intended to separate somatic, germline heterozygous, homozygous variants) was fit to this VAF distribution and a cutoff was determined such that the area under the curve of the somatic component above the cutoff was below 0.01. This gave a threshold value of 40% VAF for determining likely germline variants (LGVs). Data from 1713 TSO500 cases was then analyzed to identify LGVs. Results: The dataset of 1713 cases comprised 20.67% NSCLC, 12.61% breast, 12.49% ovarian and 10.5% colorectal cancer and < 10% of other cancer types including 2.39% pancreatic and 2.04% prostate cancer. 23.8% of ovarian, 17.14% of prostate, 12.5% of breast and 12.2% of pancreatic cancer cases had variants in BRCA1/BRCA2. Among these, 18.69% of ovarian, 17.14% of prostate, 8.33% of breast and 7.32% of pancreatic cancers had LGVs. LGVs in BRCA1/2 were also found in 3.33% of colorectal and 1.13% of NSCLC. The BRCA1/BRCA2 LGV ratio was found to be 3.0 in ovarian, 1.0 in breast and 2.0 in pancreatic but 0.2 in prostate and colorectal and 0.33 in NSCLC cancer. Adding other HBOC genes (ATM, BRCA1/2, BRIP1, CHEK2, PALB2, RAD51C, RAD51D) increased the LGV frequency to 22.43% of ovarian, 22.86% of prostate, 11.11% of breast, 9.76% of pancreatic, 4.44% of colorectal and 2.26% of NSCLC. MMR gene LGVs were seen in 1.23% of all cancers mainly in 10.71% of endometrial and 4.44% of colorectal cancers. Unlike the HBOC genes, MMR gene variants were not observed across multiple cancers and did not show a predominantly germline distribution (somatic/LGV 1:1). In contrast, the background P/LP frequency calculated from gnomAD v4 South Asian data (excluding those with conflicting labels) was 0.74% for BRCA1/2, 1.82% for the extended HBOC list and 0.5% for the MMR genes. Conclusions: Somatic testing was used to identify LGVs and determine the frequencies of HBOC and MMR genes in cancers in the Indian population. Somatic testing also identified LGVs in multiple tissues at frequencies above the background P/LP frequency for this population, suggesting a possible role for these genes in other cancers. This information underscores the need for further study and may be useful to define a clinically relevant subset of patients.

Sharp optimal recovery in the two component Gaussian mixture model

In this paper, we study the problem of clustering in the Two component Gaussian mixture model when the centers are separated by some Δ>0. We present a nonasymptotic lower bound for the corresponding minimax Hamming risk improving on existing results. We also propose an optimal, efficient and adaptive procedure that is minimax rate optimal. The rate optimality is moreover sharp in the asymptotics when the sample size goes to infinity. Our procedure is based on a variant of Lloyd’s iterations initialized by a spectral method. As a consequence of nonasymptotic results, we find a sharp phase transition for the problem of exact recovery in the Gaussian mixture model. We prove that the phase transition occurs around the critical threshold Δ¯ given by Δ¯2=σ2(1+ 1+2p nlogn)logn.

Abstract 246: Method to analyze mutational and phenotypic profiles from single cell for clonal evolution

Abstract Single cell multiomics assays targeting RNA and Protein from the same cell provide a high-resolution view of the heterogeneity of the sample. However both analytes target the phenotype of the cells and unambiguous inference that a cellular phenotype is caused by a genotype can only be achieved by their measurement from the same single cell. To address this gap, we have developed the Tapestri multi-omics workflow to analyze the DNA and Protein information from the same cell. After pre-processing the reads, cell calling is performed using DNA reads. Variant calling and filtering is carried out using DNA reads to identify the high quality genetic variants within each cell and the variant-cell matrix is then analyzed to identify clones. The Protein reads for the valid cells are log normalized. A systemic artefact resulting in random uniform amplification of antibodies is corrected for by learning the scaling factor from the read count distribution. This reduces the number of false positives. Furthermore read depth dependence of expression is corrected by identifying the correlation of the mean expression and background counts with the total reads in the cell. The mean signal and background for each cell is learned using a 2 component gaussian mixture model. Then, z-score normalization is applied to prevent high expressors from skewing further analysis. The scaled values are dimensionally reduced using PCA. A graph structure is created using KNN with weights calculated using Jaccard similarity following which community detection is employed to identify the cell types. We test this method on a model system with two donor PBMC, 2 cell lines mixture titrated at a 47:47:5:1 ratio with a 312 amplicon DNA panel and 45 plex antibody panel. We filter cells using isotype controls from the antibody panel and mixed cell signatures using the DNA panel. We can identify 4 clones using the genetic variants. We can identify multiple cell types including the major populations such as T-cells, B-cells, Monocytes and NK-cells using the phenotypic expression in the PBMC donors. Moreover, we are able to correlate the proportions of each PBMC cell type for the identified clones with that of the individual donors. Due to the corrections applied during the normalization of protein counts we are also able to accurately identify various expected expression profiles in PBMCs such as CD45RA-CD45RO and CD8-CD4 mutual exclusion, and the trimodal expression of CD4. Citation Format: Saurabh Parikh, Manimozhi Manivannan, Jacqueline Marin, Kate Thompson, Daniel Mendoza, Benjamin Schroeder, Saurabh Gulati, Shu Wang, Aik Ooi. Method to analyze mutational and phenotypic profiles from single cell for clonal evolution [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 246.

Approach to hand posture recognition based on hand shape features for human–robot interaction

Hand segmentation is the initial step for hand posture recognition. To reduce the effect of variable illumination in hand segmentation step, a new CbCr-I component Gaussian mixture model (GMM) is proposed to detect the skin region. The hand region is selected as a region of interest from the image using the skin detection technique based on the presented CbCr-I component GMM and a new adaptive threshold. A new hand shape distribution feature described in polar coordinates is proposed to extract hand contour features to solve the false recognition problem in some shape-based methods and effectively recognize the hand posture in cases when different hand postures have the same number of outstretched fingers. A multiclass support vector machine classifier is utilized to recognize the hand posture. Experiments were carried out on our data set to verify the feasibility of the proposed method. The results showed the effectiveness of the proposed approach compared with other methods.

A Spacial-Temporal Gaussian Mixture Model For Annual Average Pm2.5 Concentration Analysis

PM2.5 is a major air pollutant which has a high probability to cause many serious cardiopulmonary diseases, such as asthma, lung cancer, trachea cancer, bronchus cancer, etc. Up to 2014, a World Health Organization (WHO) air quality model confirmed that 92% of the population in the world lived in areas where air quality levels exceeded WHO limits (i.e., 10 µg/m3). This indicates that PM2.5 is still one of the most serious world-wide problems, and monitoring PM2.5 concentrations is extremely necessary. In this paper, we proposed a easy and flexible spatial-temporal Gaussian mixture model to analyze annual average PM2.5 concentrations. Because of the bimodal distribution of PM2.5 concentrations, we decided for a two- component Gaussian mixture model with county-year-level spatial-temporal random effects. A Markov Chain Monte Carlo (MCMC) algorithm is used to estimating model parameters.

Two-Level Classification in Determining the Age and Gender Group of a Speaker

In this study, the classification of the speakers according to age and gender was discussed. Age and gender classes were first examined separately, and then by combining these classes a classification with a total of 7 classes was made. Speech signals represented by Mel-Frequency Cepstral Coefficients (MFCC) and delta parameters were converted into Gaussian Mixture Model (GMM) mean supervectors and classified with a Support Vector Machine (SVM). While the GMM mean supervectors were formed according to the Maximum-a-posteriori (MAP) adaptive GMM-Universal Background Model (UBM) configuration, the number of components was changed from 16 to 512, and the optimum number of components was decided. Gender classification accuracy of the system developed using aGender dataset was measured as 99.02% for two classes and 92.58% for three classes and age group classification accuracy was measured as 67.03% for female and 63.79% for male. In the classification of age and gender classes together in one step, an accuracy of 61.46% was obtained. In the study, a two-level approach was proposed for classifying age and gender classes together. According to this approach, the speakers were first divided into three classes as child, male and female, then males and females were classified according to their age groups and thus a 7-class classification was realized. This two-level approach was increased the accuracy of the classification in all other cases except when 32-component GMMs were used. While the highest improvement of 2.45% was achieved with 64 component GMMs, an improvement of 0.79 was achieved with 256 component GMMs.

Ses Telleri Görüntülerinde Otomatik Piksel Tabanlı Sınıflandırma için Performans Ölçütlerinin İncelenmesi

In recently years, thanks to the opportunities brought by the developing technology, high-speed images of the vocal cords have been started to widely use in detection of problems with the speech system and analysis of speech. These high-speed images contain detailed information about the vibration of the speaker's vocal cords. However, considering the size of the image data, it does not seem possible to manually process these images. For this reason, glottis detection and segmentation from vocal cord images has become popular with the development of automatic image processing algorithms in recent years. Unlike the other literature studies, in this study, the accuracy, precision (sensitivity), recall, F1-score and equal error rate performance criteria are examined used to automatically classify vocal cord images based on pixels. In addition to this, deep artificial neural network, that pixel classification based model in the literature, has been compared with the newly proposed model Gaussian Mixture Model. 3000 high speed endoscopic camera images manually segmented with dimensions 256x256 pixels were used to generate training, development and evaluation data sets of randomly. As a result of the studies conducted with the validation and evaluation sets of models trained with the data set, the accuracy, precision, recall and F1 score criteria, which are commonly used in binary classification, changed only by 1% from model to model. And this result has shown that other performance metrics are not as effective as equal error rate that reflecting the system 22% change in the same situation. As a result of this study, even if the accuracy values of the systems remain the same, equal error rate differences may change, therefore it has been shown that overfitted systems can be predicted more accurately. Comparing the models proposed with the based system, the proposed system gave the best result for all performance criteria using the 4096 component Gaussian Mixture Model, and it is showed a performance improvement of 22% for the equal error rate in the evaluation set.

Bayesian classification of falls risk

BackgroundPrior research in falls risk prediction often relies on qualitative and/or clinical methods. There are two challenges with these methods. First, qualitative methods typically use falls history to determine falls risk. Second, clinical methods do not quantify the uncertainty in the classification decision. In this paper, we propose using Bayesian classification to predict falls risk using vectors of gait variables shown to contribute to falls risk. Research Questions(1) Using a vector of risk ratios for specific gait variables shown to contribute to falls risk, how can older adults be classified as low or high falls risk? and (2) how can the uncertainty in the classifier decision be quantified when using a vector of gait variables? MethodsUsing a pressure sensitive walkway, biomechanical measurements of gait were collected from 854 adults over the age of 65. In our method, we first determine low and high falls risk labels for vectors of risk ratios using the k-means algorithm. Next, the posterior probability of low or high falls risk class membership is obtained from a two component Gaussian mixture model (GMM) of gait vectors, which enables risk assessment directly from the underlying biomechanics. We classify the gait vectors using a threshold based on Youden's J statistic. ResultsThrough a Monte Carlo simulation and an analysis of the receiver operating characteristic (ROC), we demonstrate that our Bayesian classifier, when compared to the k-means falls risk labels, achieves an accuracy greater than 96% at predicting low or high falls risk. SignificanceOur analysis indicates that our approach based on a Bayesian framework and an individual's underlying biomechanics can predict falls risk while quantifying uncertainty in the classification decision.

Sparse broadband time varying acoustic response modeling and estimation from an undersampled vertical array with application to underwater communications.

Sparse broadband time varying acoustic response functions associated with horizontally stratified environments can be well estimated from an undersampled vertical array. The sparsity of the arrivals here is captured by means of a two component Gaussian mixture model. The implied posterior expectation of the bifrequency function serves to attenuate low amplitude non-coherent arrivals while leaving coherent arrivals at the array unchanged. This estimate of the acoustic response specifies a time varying increment operator that unravels the time varying motion between the source and receiver array allowing for a further sparsification of the response. This additional improved sparsity allows for improved coherent multipath combining at the receiver. The model is applied to a three element vertical array during a set of broadband acoustic observations in the shallow waters of Buzzard's Bay, MA. M-ary orthogonal spread spectrum acoustic transmissions at 32.5 kHz center frequency and 25 kHz bandwidth are considered. With M = 4, throughput rates of 95 bps are tested. At 2 km and at a receive signal-to-noise ratio per element of -15.5 dB, a 3 element undersampled vertical array achieved bit error rates less than 10-5.

Determining air-conditioning usage patterns in Singapore from distributed, portable sensors

In Singapore, a large portion of residential electricity consumption is typically used for space cooling. The operation patterns and respective energy consumption of residential split system air-conditioning (AC) units depend on residents’ behavior and are not well known. Singapore's National Science Experiment (NSE) is a nation-wide experiment that involves students carrying a small sensor device to collect environmental data. Over 43,000 students participated in 2015, and the experiment is still ongoing. In this study, we use the Expectation-Maximization algorithm to fit a two component Gaussian mixture model to measured data of temperature, humidity, and pressure to infer patterns of AC exposure. The daily probability profile of AC exposure of Singaporean students was determined. The results can be used as a basic dataset for thermal comfort and energy consumption behavior studies. Derived patterns of AC usage could be used in building energy simulation.

Identification Features Analysis in Speech Data Using Gmm-Ubm Speaker Verification System

This paper is devoted to feature selection and evaluation in an automatic text-independent speaker verification task. In order to solve this problem a speaker verification system based on the Gaussian mixture model and the universal background model (GMM-UBM system) was used. The application sphere and challenges of modern systems of automatic speaker identification were considered. Overview of the modern speaker recognition methods and main speech features used in speaker identification is provided. Features extraction process used in this article was examined. Reviewed speech features were used for speaker verification including mel-cepstral coefficients (MFCC), line spectral pairs (LSP), perceptual linear prediction cepstral coefficients (PLP), short-term energy, formant frequencies, fundamental frequency, voicing probability, zero crossing rate (ZCR), jitter and shimmer. The experimental evaluation of the GMM-UBM system using different speech features was conducted on a 50 speaker set and a result is presented. Feature selection was done using the genetic algorithm and the greedy adding and deleting algorithm. Equal error rate (EER) equals 0,579 % when using 256 component Gaussian mixture model and the obtained feature vector. Comparing to standard 14 MFCC vector, 42,1 % of EER improvement was acquired.

Parametric Reconstruction of Diffuse Optical Tomography Using Gaussian Mixture Model and Genetic Algorithm

The inverse problem of diffused optical tomography is highly ill-posed as well as often nonunique and unstable. The ill-posedness can be reduced by reducing the number of unknown values (e.g., adding soft or hard-prior constraints). In this paper, this has been achieved by parameterization of the tissue optical properties. The optical parameters of the tissue have been modeled by a Gaussian mixture model (GMM) of eight components, and the image is reconstructed by estimating the parameters of the mixture model. Therefore, the number of unknowns is drastically reduced while maintaining some amount of generality by using an 8 component GMM. Unlike conventional GMM, the parameters are estimated using the Genetic algorithm in a multi-step predictor-corrector model. A new cost function is proposed using mean square error and asymmetry measurement of detected intensity. Object centroid error, structural similarity index, maximum error and normalized root mean squared error of reconstructed absorption coefficients are evaluated for performance assessment and comparison. Extensive numerical and experimental results have been illustrated to show the potential of the proposed method. Use of CW (continuous wave) measurement and single imaging modality make the system cost effective and portable.

Interpolation on the manifold of K component GMMs.

Probability density functions (PDFs) are fundamental objects in mathematics with numerous applications in computer vision, machine learning and medical imaging. The feasibility of basic operations such as computing the distance between two PDFs and estimating a mean of a set of PDFs is a direct function of the representation we choose to work with. In this paper, we study the Gaussian mixture model (GMM) representation of the PDFs motivated by its numerous attractive features. (1) GMMs are arguably more interpretable than, say, square root parameterizations (2) the model complexity can be explicitly controlled by the number of components and (3) they are already widely used in many applications. The main contributions of this paper are numerical algorithms to enable basic operations on such objects that strictly respect their underlying geometry. For instance, when operating with a set of K component GMMs, a first order expectation is that the result of simple operations like interpolation and averaging should provide an object that is also a K component GMM. The literature provides very little guidance on enforcing such requirements systematically. It turns out that these tasks are important internal modules for analysis and processing of a field of ensemble average propagators (EAPs), common in diffusion weighted magnetic resonance imaging. We provide proof of principle experiments showing how the proposed algorithms for interpolation can facilitate statistical analysis of such data, essential to many neuroimaging studies. Separately, we also derive interesting connections of our algorithm with functional spaces of Gaussians, that may be of independent interest.

Language identification using multi-core processors

Graphics processing units (GPUs) provide substantial processing power for little cost. We explore the application of GPUs to speech pattern processing, using language identification (LID) to demonstrate their benefits. Realization of the full potential of GPUs requires both effective coding of predetermined algorithms, and, if there is a choice, selection of the algorithm or technique for a specific function that is most able to exploit the GPU. We demonstrate these principles using the NIST LRE 2003 standard LID task, a batch processing task which involves the analysis of over 600h of speech. We focus on two parts of the system, namely the acoustic classifier, which is based on a 2048 component Gaussian Mixture Model (GMM), and acoustic feature extraction. In the case of the latter we compare a conventional FFT-based analysis with IIR and FIR filter banks, both in terms of their ability to exploit the GPU architecture and LID performance. With no increase in error rate our GPU based system, with an FIR-based front-end, completes the NIST LRE 2003 task in 16h, compared with 180h for the conventional FFT-based system on a standard CPU (a speed up factor of more than 11). This includes a 61% decrease in front-end processing time. In the GPU implementation, front-end processing accounts for 8% and 10% of the total computing times during training and recognition, respectively. Hence the reduction in front-end processing achieved in the GPU implementation is significant.

Generalized Time-Series Active Search With Kullback–Leibler Distance for Audio Fingerprinting

In this letter, a new audio fingerprinting approach is presented. We investigate to improve robustness by more precise statistical fingerprint modeling with common component Gaussian mixture models (CCGMMs) and Kullback-Leibler (KL) distance, which is more suitable to measure the dissimilarity between two probabilistic models. To address the resulting complexity, generalized time-series active search is proposed, which supports a wide variety of distance measures between two CCGMMs, including L1, L2, KL, etc. Experiments show that the new approach with KL distance increases robustness to distortions (including low-quality MP3 compression, small room echo, and play-and-record) while achieving efficient search