Vec2image: an explainable artificial intelligence model for the feature representation and classification of high-dimensional biological data by vector-to-image conversion
Abstract:Feature representation and discriminative learning are proven models and technologies in artificial intelligence fields; however, major challenges for machine learning on large biological datasets are learning an effective model with mechanistical explanation on the model determination and prediction. To satisfy such demands, we developed Vec2image, an explainable convolutional neural network framework for characterizing the feature engineering, feature selection and classifier training that is mainly based on… Show more
“…S7 clearly demonstrate the superiority of the proposed approach over the existing methods. Noteworthy, the performance of our technique drops only by 2–3% for the study with 30% of total data for training, whereas the decrease in classification accuracy of all other techniques (except ACTINN 27 and Vec2image 28 ) is at least 7%. Fig.…”
Section: Resultsmentioning
confidence: 81%
“…In these methods, however, an image is usually made heuristically by projecting the HD data onto a 2D plane without any explicit constraint(s) on the spatial locations of the genes. As an example, In vec2image 28 , t -SNE (or other) embedding method is used to create the images. As a result, similar genes get clustered under the assumption of t-distribution without explicit constraints on their spatial positions for maximizing entropy.…”
Section: Discussionmentioning
confidence: 99%
“…Furthermore, the concept of reconfiguring tabulated data into an image is quite general and we envision that the technique will find valuable applications in dealing with HD data across disciplines, such as healthcare, finance, and physical sciences. We note that there were a few attempts in the literature to convert gene expression data into images for deep learning 28,[62][63][64] . In these methods, however, an image is usually made heuristically by projecting the HD data onto a 2D plane without any explicit constraint(s) on the spatial locations of the genes.…”
Remarkable advances in single cell genomics have presented unique challenges and opportunities for interrogating a wealth of biomedical inquiries. High dimensional genomic data are inherently complex because of intertwined relationships among the genes. Existing methods, including emerging deep learning-based approaches, do not consider the underlying biological characteristics during data processing, which greatly compromises the performance of data analysis and hinders the maximal utilization of state-of-the-art genomic techniques. In this work, we develop an entropy-based cartography strategy to contrive the high dimensional gene expression data into a configured image format, referred to as genomap, with explicit integration of the genomic interactions. This unique cartography casts the gene-gene interactions into the spatial configuration of genomaps and enables us to extract the deep genomic interaction features and discover underlying discriminative patterns of the data. We show that, for a wide variety of applications (cell clustering and recognition, gene signature extraction, single cell data integration, cellular trajectory analysis, dimensionality reduction, and visualization), the proposed approach drastically improves the accuracies of data analyses as compared to the state-of-the-art techniques.
“…S7 clearly demonstrate the superiority of the proposed approach over the existing methods. Noteworthy, the performance of our technique drops only by 2–3% for the study with 30% of total data for training, whereas the decrease in classification accuracy of all other techniques (except ACTINN 27 and Vec2image 28 ) is at least 7%. Fig.…”
Section: Resultsmentioning
confidence: 81%
“…In these methods, however, an image is usually made heuristically by projecting the HD data onto a 2D plane without any explicit constraint(s) on the spatial locations of the genes. As an example, In vec2image 28 , t -SNE (or other) embedding method is used to create the images. As a result, similar genes get clustered under the assumption of t-distribution without explicit constraints on their spatial positions for maximizing entropy.…”
Section: Discussionmentioning
confidence: 99%
“…Furthermore, the concept of reconfiguring tabulated data into an image is quite general and we envision that the technique will find valuable applications in dealing with HD data across disciplines, such as healthcare, finance, and physical sciences. We note that there were a few attempts in the literature to convert gene expression data into images for deep learning 28,[62][63][64] . In these methods, however, an image is usually made heuristically by projecting the HD data onto a 2D plane without any explicit constraint(s) on the spatial locations of the genes.…”
Remarkable advances in single cell genomics have presented unique challenges and opportunities for interrogating a wealth of biomedical inquiries. High dimensional genomic data are inherently complex because of intertwined relationships among the genes. Existing methods, including emerging deep learning-based approaches, do not consider the underlying biological characteristics during data processing, which greatly compromises the performance of data analysis and hinders the maximal utilization of state-of-the-art genomic techniques. In this work, we develop an entropy-based cartography strategy to contrive the high dimensional gene expression data into a configured image format, referred to as genomap, with explicit integration of the genomic interactions. This unique cartography casts the gene-gene interactions into the spatial configuration of genomaps and enables us to extract the deep genomic interaction features and discover underlying discriminative patterns of the data. We show that, for a wide variety of applications (cell clustering and recognition, gene signature extraction, single cell data integration, cellular trajectory analysis, dimensionality reduction, and visualization), the proposed approach drastically improves the accuracies of data analyses as compared to the state-of-the-art techniques.
“…Several studies have proposed methods for projecting n feature vectors into 2D space by reducing them to p(<n) dimensions using dimension reduction methodology [34][35][36]. When dimensionality reduction is performed, high-dimensional features are displayed in a space that can be visually recognized by humans, so it has the advantage of intuitively helping to understand data and reducing the number of features.…”
Section: Studies On Converting Tabular Data Into Imagesmentioning
Since demand is influenced by a wide variety of causes, it is necessary to decompose the explanatory variables into different levels, extract their relationships effectively, and reflect them in the forecast. In particular, this contextual information can be very useful in demand forecasting with large demand volatility or intermittent demand patterns. Convolutional neural networks (CNNs) have been successfully used in many fields where important information in data is represented by images. CNNs are powerful because they accept samples as images and use adjacent voxel sets to integrate multi-dimensional important information and learn important features. On the other hand, although the demand-forecasting model has been improved, the input data is still limited in its tabular form and is not suitable for CNN modeling. In this study, we propose a Tab2vox neural architecture search (NAS) model as a method to convert a high-dimensional tabular sample into a well-formed 3D voxel image and use it in a 3D CNN network. For each image representation, the 3D CNN forecasting model proposed from the Tab2vox framework showed superior performance, compared to the existing time series and machine learning techniques using tabular data, and the latest image transformation studies.
“…The DeepInsight method pioneered a strategy by converting non-image data to image form and then processing it to CNN for classification for various kinds of datasets. It has been widely used in various fields such as in cancer research [10][11][12], viral infections [13], sparse data [14], power energy [15], business and manufacturing [16], time-series data [17][18][19], traffic cash analysis [20], human activity recognition [21], feature representation [22], intrusion detection [23], spine surgery [24] and HVAC fault diagnosis [25]. Moreover, DeepInsight was a component in the Kaggle.com competition hosted by MIT and Harvard University that secured rank1 on the leaderboard [26].…”
Modern oncology offers a wide range of treatments and therefore choosing the best option for particular patient is very important for optimal outcomes. Multi-omics profiling in combination with AI-based predictive models have great potential for streamlining these treatment decisions. However, these encouraging developments continue to be hampered by very high dimensionality of the datasets in combination with insufficiently large numbers of annotated samples. In this study, we propose a novel deep learning-based method to predict patient-specific anticancer drug response from three types of multiomics data. The proposed DeepInsight-3D approach relies on structured data-to-image conversion that then allows use of convolutional neural networks, which are particularly robust to high dimensionality of the inputs while retaining capabilities to model highly complex relationships between variables. Of particular note, we demonstrate that in this formalism additional channels of an image can be effectively used to accommodate data from different omics layers while explicitly encoding the connection between them. DeepInsight-3D was able to outperform two other state-of-the-art methods proposed for this task. These advances can facilitate the development of better personalized treatment strategies for different cancers in the future.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
