scholarly journals The impact of different negative training data on regulatory sequence predictions

PLoS ONE ◽  
2020 ◽  
Vol 15 (12) ◽  
pp. e0237412
Author(s):  
Louisa-Marie Krützfeldt ◽  
Max Schubach ◽  
Martin Kircher
Keyword(s):  
Model Performance ◽  
Training Data ◽  
Training Dataset ◽  
Support Vector ◽  
Regulatory Sequence ◽  
Open Chromatin ◽  
Regulatory Sequences ◽  
Cell Type ◽  
The Impact ◽  
Negative Training

Regulatory regions, like promoters and enhancers, cover an estimated 5–15% of the human genome. Changes to these sequences are thought to underlie much of human phenotypic variation and a substantial proportion of genetic causes of disease. However, our understanding of their functional encoding in DNA is still very limited. Applying machine or deep learning methods can shed light on this encoding and gapped k-mer support vector machines (gkm-SVMs) or convolutional neural networks (CNNs) are commonly trained on putative regulatory sequences. Here, we investigate the impact of negative sequence selection on model performance. By training gkm-SVM and CNN models on open chromatin data and corresponding negative training dataset, both learners and two approaches for negative training data are compared. Negative sets use either genomic background sequences or sequence shuffles of the positive sequences. Model performance was evaluated on three different tasks: predicting elements active in a cell-type, predicting cell-type specific elements, and predicting elements' relative activity as measured from independent experimental data. Our results indicate strong effects of the negative training data, with genomic backgrounds showing overall best results. Specifically, models trained on highly shuffled sequences perform worse on the complex tasks of tissue-specific activity and quantitative activity prediction, and seem to learn features of artificial sequences rather than regulatory activity. Further, we observe that insufficient matching of genomic background sequences results in model biases. While CNNs achieved and exceeded the performance of gkm-SVMs for larger training datasets, gkm-SVMs gave robust and best results for typical training dataset sizes without the need of hyperparameter optimization.

scholarly journals The impact of different negative training data on regulatory sequence predictions

2020 ◽  
Author(s):  
Louisa-Marie Krützfeldt ◽  
Max Schubach ◽  
Martin Kircher
Keyword(s):  
Model Performance ◽  
Training Data ◽  
Training Dataset ◽  
Support Vector ◽  
Regulatory Sequence ◽  
Open Chromatin ◽  
Regulatory Sequences ◽  
Cell Type ◽  
The Impact ◽  
Negative Training

AbstractRegulatory regions, like promoters and enhancers, cover an estimated 5-15% of the human genome. Changes to these sequences are thought to underlie much of human phenotypic variation and a substantial proportion of genetic causes of disease. However, our understanding of their functional encoding in DNA is still very limited. Applying machine or deep learning methods can shed light on this encoding and gapped k-mer support vector machines (gkm-SVMs) or convolutional neural networks (CNNs) are commonly trained on putative regulatory sequences.Here, we investigate the impact of negative sequence selection on model performance. By training gkm-SVM and CNN models on open chromatin data and corresponding negative training dataset, both learners and two approaches for negative training data are compared. Negative sets use either genomic background sequences or sequence shuffles of the positive sequences. Model performance was evaluated on three different tasks: predicting elements active in a cell-type, predicting cell-type specific elements, and predicting elements’ relative activity as measured from independent experimental data.Our results indicate strong effects of the negative training data, with genomic backgrounds showing overall best results. Specifically, models trained on highly shuffled sequences perform worse on the complex tasks of tissue-specific activity and quantitative activity prediction, and seem to learn features of artificial sequences rather than regulatory activity. Further, we observe that insufficient matching of genomic background sequences results in model biases. While CNNs achieved and exceeded the performance of gkm-SVMs for larger training datasets, gkm-SVMs gave robust and best results for typical training dataset sizes without the need of hyperparameter optimization.

scholarly journals DANNP: an efficient artificial neural network pruning tool

2017 ◽  
Vol 3 ◽  
pp. e137 ◽  
Author(s):  
Mona Alshahrani ◽  
Othman Soufan ◽  
Arturo Magana-Mora ◽  
Vladimir B. Bajic
Keyword(s):  
Neural Network ◽  
State Of The Art ◽  
Model Performance ◽  
Training Data ◽  
Classification Problems ◽  
Link Type ◽  
On Line ◽  
Pruning Algorithms ◽  
Artificial Neural ◽  
The Impact

Background Artificial neural networks (ANNs) are a robust class of machine learning models and are a frequent choice for solving classification problems. However, determining the structure of the ANNs is not trivial as a large number of weights (connection links) may lead to overfitting the training data. Although several ANN pruning algorithms have been proposed for the simplification of ANNs, these algorithms are not able to efficiently cope with intricate ANN structures required for complex classification problems. Methods We developed DANNP, a web-based tool, that implements parallelized versions of several ANN pruning algorithms. The DANNP tool uses a modified version of the Fast Compressed Neural Network software implemented in C++ to considerably enhance the running time of the ANN pruning algorithms we implemented. In addition to the performance evaluation of the pruned ANNs, we systematically compared the set of features that remained in the pruned ANN with those obtained by different state-of-the-art feature selection (FS) methods. Results Although the ANN pruning algorithms are not entirely parallelizable, DANNP was able to speed up the ANN pruning up to eight times on a 32-core machine, compared to the serial implementations. To assess the impact of the ANN pruning by DANNP tool, we used 16 datasets from different domains. In eight out of the 16 datasets, DANNP significantly reduced the number of weights by 70%–99%, while maintaining a competitive or better model performance compared to the unpruned ANN. Finally, we used a naïve Bayes classifier derived with the features selected as a byproduct of the ANN pruning and demonstrated that its accuracy is comparable to those obtained by the classifiers trained with the features selected by several state-of-the-art FS methods. The FS ranking methodology proposed in this study allows the users to identify the most discriminant features of the problem at hand. To the best of our knowledge, DANNP (publicly available at www.cbrc.kaust.edu.sa/dannp) is the only available and on-line accessible tool that provides multiple parallelized ANN pruning options. Datasets and DANNP code can be obtained at www.cbrc.kaust.edu.sa/dannp/data.php and https://doi.org/10.5281/zenodo.1001086.

scholarly journals Development of an ANN-Based Urban Flood Alert Criteria Prediction Model and the Impact of Training Data Augmentation

2021 ◽  
Vol 21 (6) ◽  
pp. 257-264
Author(s):  
Hoseon Kang ◽  
Jaewoong Cho ◽  
Hanseung Lee ◽  
Jeonggeun Hwang ◽  
Hyejin Moon
Keyword(s):  
Artificial Intelligence ◽  
Data Augmentation ◽  
Model Performance ◽  
Fuzzy Model ◽  
Flood Damage ◽  
Training Data ◽  
Ann Model ◽  
Urban Flood ◽  
Flood Alert ◽  
The Impact

Urban flooding occurs during heavy rains of short duration, so quick and accurate warnings of the danger of inundation are required. Previous research proposed methods to estimate statistics-based urban flood alert criteria based on flood damage records and rainfall data, and developed a Neuro-Fuzzy model for predicting appropriate flood alert criteria. A variety of artificial intelligence algorithms have been applied to the prediction of the urban flood alert criteria, and their usage and predictive precision have been enhanced with the recent development of artificial intelligence. Therefore, this study predicted flood alert criteria and analyzed the effect of applying the technique to augmentation training data using the Artificial Neural Network (ANN) algorithm. The predictive performance of the ANN model was RMSE 3.39-9.80 mm, and the model performance with the extension of training data was RMSE 1.08-6.88 mm, indicating that performance was improved by 29.8-82.6%.

scholarly journals Improving generalization of deep learning models for diagnostic pathology by increasing variability in training data: experiments on osteosarcoma subtypes

2020 ◽  
Author(s):  
Haiming Tang ◽  
Nanfei Sun ◽  
Steven Shen
Keyword(s):  
Deep Learning ◽  
Model Performance ◽  
High Variability ◽  
Training Data ◽  
Classification Model ◽  
Training Dataset ◽  
Learning Models ◽  
Diagnostic Pathology ◽  
Model Generalization ◽  
Histopathological Images

Artificial intelligence (AI) has an emerging progress in diagnostic pathology. A large number of studies of applying deep learning models to histopathological images have been published in recent years. While many studies claim high accuracies, they may fall into the pitfalls of overfitting and lack of generalization due to the high variability of the histopathological images. We use the example of Osteosarcoma to illustrate the pitfalls and how the addition of model input variability can help improve model performance. We use the publicly available osteosarcoma dataset to retrain a previously published classification model for osteosarcoma. We partition the same set of images into the training and testing datasets differently than the original study: the test dataset consists of images from one patient while the training dataset consists images of all other patients. The performance of the model on the test set using the new partition schema declines dramatically, indicating a lack of model generalization and overfitting.We also show the influence of training data variability on model performance by collecting a minimal dataset of 10 osteosarcoma subtypes as well as benign tissues and benign bone tumors of differentiation. We show the additions of more and more subtypes into the training data step by step under the same model schema yield a series of coherent models with increasing performances. In conclusion, we bring forward data preprocessing and collection tactics for histopathological images of high variability to avoid the pitfalls of overfitting and build deep learning models of higher generalization abilities.

scholarly journals Performance of three machine learning algorithms for predicting soil organic carbon in German agricultural soil

2021 ◽  
Author(s):  
Ali Sakhaee ◽  
Anika Gebauer ◽  
Mareike Ließ ◽  
Axel Don
Keyword(s):  
Machine Learning ◽  
Organic Carbon ◽  
Soil Organic Carbon ◽  
Agricultural Soil ◽  
Learning Algorithms ◽  
Model Performance ◽  
Machine Learning Algorithms ◽  
Support Vector ◽  
Organic Soils ◽  
The Impact

Abstract. Soil organic carbon (SOC), as the largest terrestrial carbon pool, has the potential to influence climate change and mitigation, and consequently SOC monitoring is important in the frameworks of different international treaties. There is therefore a need for high resolution SOC maps. Machine learning (ML) offers new opportunities to do this due to its capability for data mining of large datasets. The aim of this study, therefore, was to test three commonly used algorithms in digital soil mapping – random forest (RF), boosted regression trees (BRT) and support vector machine for regression (SVR) – on the first German Agricultural Soil Inventory to model agricultural topsoil SOC content. Nested cross-validation was implemented for model evaluation and parameter tuning. Moreover, grid search and differential evolution algorithm were applied to ensure that each algorithm was tuned and optimised suitably. The SOC content of the German Agricultural Soil Inventory was highly variable, ranging from 4 g kg−1 to 480 g kg−1. However, only 4 % of all soils contained more than 87 g kg−1 SOC and were considered organic or degraded organic soils. The results show that SVR provided the best performance with RMSE of 32 g kg−1 when the algorithms were trained on the full dataset. However, the average RMSE of all algorithms decreased by 34 % when mineral and organic soils were modeled separately, with the best result from SVR with RMSE of 21 g kg−1. Model performance is often limited by the size and quality of the available soil dataset for calibration and validation. Therefore, the impact of enlarging the training data was tested by including 1223 data points from the European Land Use/Land Cover Area Frame Survey for agricultural sites in Germany. The model performance was enhanced for maximum 1 % for mineral soils and 2 % for organic soils. Despite the capability of machine learning algorithms in general, and particularly SVR, in modelling SOC on a national scale, the study showed that the most important to improve the model performance was separate modelling of mineral and organic soils.

Long-term Runoff Forecasting Models Based on the Teleconnection coupled with Machine Learning

2020 ◽  
Author(s):  
Teng Zhang ◽  
Zhongjing Wang ◽  
Zixiong Zhang
Keyword(s):  
Support Vector Machine ◽  
Water Resources ◽  
Model Performance ◽  
Support Vector ◽  
Wet Season ◽  
Model Based ◽  
Runoff Forecasting ◽  
Monthly Runoff ◽  
Highly Correlated ◽  
The Impact

<p>Runoff forecast with high precision is important for the efficient utilization of water resources and regional sustainable development, especially in the arid area. The monthly runoff of Changmabao (CMB) station has an upwards trend and an abrupt point in 1998. The impact factor analysis shows that it is highly correlated with the current precipitation and temperature in the wet season while the previous runoff and previous global land temperature in the dry season. Three models including the time-series decomposition model, the model based on teleconnection coupled with the support vector machine, and the model based on teleconnection coupled with the artificial neural network are used to predict the runoff of CMB station. An indicator β is constructed with the correlation coefficient (R) and mean relative deviation (rBias) to evaluate the model performance more conveniently and intuitively. The results suggest that the model based on teleconnection coupled with the support vector machine preforms best. This forecasting method could be applied to the management and dispatch of water resources in arid areas.</p>

scholarly journals Statistical Analysis for Twitter Spam Detection

2019 ◽  
pp. 624-629
Author(s):  
Ganesh Udge ◽  
Mahesh Mohite ◽  
Shubhankar Bendre ◽  
Yogeshwar Birnagal ◽  
Disha Wankhede
Keyword(s):  
Online Social Networks ◽  
Binary Classification ◽  
Classification Model ◽  
Training Dataset ◽  
Support Vector ◽  
Spam Detection ◽  
Data Sampling ◽  
Quality Of Data ◽  
Related Factors ◽  
The Impact

The spreading and learning of new discoveries and information is made available using current online social networks. In Recent days, the solutions may be irrelevant to the actual content; also termed as attacks in the layman’s term such attacks are been performed on Twitter as well and called as Twitter spammers. The quality of data is being compromised by addition of malicious and harmful information using URL, bio, emoticons, audio, images/videos & hash-tags through different accounts by exchanging tweets, personal messages (Direct Message’s) & re-tweets. Misleading sites may be linked with the malicious links which may affect adverse effects on the user and also interfere in their decision making processes. To improve user-experience from the spammers attacks, the training twitter dataset are applied and then by extracting and using the 12 lightweight features like user’s age, number of followers, count of tweets and re-tweets, etc. are used to distinguish the spam from non-spam. For enhancing the performance, the discretization of the function is important for transmission of spam detection between tweets. Our system creates classification model for Spam detection which includes binary classification and automatic learning algorithms viz. Naïve Bayes classifier or Support Vector Machine classifier which understands the behaviour of the model. The system will categorize the tweets from datasets into Spam and Non-spam classes and provide the user’s feed with only the relevant information. The system will report the impact of data-related factors such as relationship between spam and non-spam tweets, size of training dataset, data sampling and detection performance. The proposed system’s function is detection and analysis of the simple and variable twitter spam over time. The spam detection is a major challenge for the system and shortens the gap between performance appraisals and focuses primarily on data, features and patterns to identify real user and informing it about the spam tweets along with the performance statistics. The work is to detect spammed tweets in real time, since the new tweets may show patterns and this will help for training and updating dataset and in knowledge base.

scholarly journals Addiction-associated genetic variants implicate brain cell type- and region-specific cis-regulatory elements in addiction neurobiology

2020 ◽  
Author(s):  
Chaitanya Srinivasan ◽  
BaDoi N. Phan ◽  
Alyssa J. Lawler ◽  
Easwaran Ramamurthy ◽  
Michael Kleyman ◽  
...  
Keyword(s):  
Genetic Variants ◽  
Cell Types ◽  
Regulatory Elements ◽  
Brain Regions ◽  
Open Chromatin ◽  
Genome Wide Association Studies ◽  
Cell Type ◽  
Coding Regions ◽  
Cell Type Specific ◽  
The Impact

ABSTRACTRecent large genome-wide association studies (GWAS) have identified multiple confident risk loci linked to addiction-associated behavioral traits. Genetic variants linked to addiction-associated traits lie largely in non-coding regions of the genome, likely disrupting cis-regulatory element (CRE) function. CREs tend to be highly cell type-specific and may contribute to the functional development of the neural circuits underlying addiction. Yet, a systematic approach for predicting the impact of risk variants on the CREs of specific cell populations is lacking. To dissect the cell types and brain regions underlying addiction-associated traits, we applied LD score regression to compare GWAS to genomic regions collected from human and mouse assays for open chromatin, which is associated with CRE activity. We found enrichment of addiction-associated variants in putative regulatory elements marked by open chromatin in neuronal (NeuN+) nuclei collected from multiple prefrontal cortical areas and striatal regions known to play major roles in reward and addiction. To further dissect the cell type-specific basis of addiction-associated traits, we also identified enrichments in human orthologs of open chromatin regions of mouse neuron subtypes: cortical excitatory, PV, D1, and D2. Lastly, we developed machine learning models from mouse cell type-specific regions of open chromatin to further dissect human NeuN+ open chromatin regions into cortical excitatory or striatal D1 and D2 neurons and predict the functional impact of addiction-associated genetic variants. Our results suggest that different neuron subtypes within the reward system play distinct roles in the variety of traits that contribute to addiction.Significance StatementOur study on cell types and brain regions contributing to heritability of addiction-associated traits suggests that the conserved non-coding regions within cortical excitatory and striatal medium spiny neurons contribute to genetic predisposition for nicotine, alcohol, and cannabis use behaviors. This computational framework can flexibly integrate epigenomic data across species to screen for putative causal variants in a cell type- and tissue-specific manner across numerous complex traits.

scholarly journals Data based predictive models for odor perception

2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Rinu Chacko ◽  
Deepak Jain ◽  
Manasi Patwardhan ◽  
Abhishek Puri ◽  
Shirish Karande ◽  
...  
Keyword(s):  
Machine Learning ◽  
Data Analytics ◽  
Model Performance ◽  
Structural Features ◽  
Machine Learning Algorithms ◽  
Gradient Boosting ◽  
Support Vector ◽  
Structure Property ◽  
Odor Perception ◽  
The Impact

Abstract Machine learning and data analytics are being increasingly used for quantitative structure property relation (QSPR) applications in the chemical domain where the traditional Edisonian approach towards knowledge-discovery have not been fruitful. The perception of odorant stimuli is one such application as olfaction is the least understood among all the other senses. In this study, we employ machine learning based algorithms and data analytics to address the efficacy of using a data-driven approach to predict the perceptual attributes of an odorant namely the odorant characters (OC) of “sweet” and “musky”. We first analyze a psychophysical dataset containing perceptual ratings of 55 subjects to reveal patterns in the ratings given by subjects. We then use the data to train several machine learning algorithms such as random forest, gradient boosting and support vector machine for prediction of the odor characters and report the structural features correlating well with the odor characters based on the optimal model. Furthermore, we analyze the impact of the data quality on the performance of the models by comparing the semantic descriptors generally associated with a given odorant to its perception by majority of the subjects. The study presents a methodology for developing models for odor perception and provides insights on the perception of odorants by untrained human subjects and the effect of the inherent bias in the perception data on the model performance. The models and methodology developed here could be used for predicting odor characters of new odorants.

scholarly journals Hoax News Classification using Machine Learning Algorithms

2019 ◽  
Vol 9 (2) ◽  
pp. 3938-3944
Keyword(s):  
Machine Learning ◽  
Social Media ◽  
Learning Algorithm ◽  
Detection System ◽  
Machine Learning Algorithms ◽  
Training Data ◽  
Stochastic Gradient Descent ◽  
Support Vector ◽  
The Impact ◽  
F Measure

Hoax news on social media has had a dramatic effect on our society in recent years. The impact of hoax news felt by many people, anxiety, financial loss, and loss of the right name. Therefore we need a detection system that can help reduce hoax news on social media. Hoax news classification is one of the stages in the construction of a hoax news detection system, and this unsupervised learning algorithm becomes a method for creating hoax news datasets, machine learning tools for data processing, and text processing for detecting data. The next will produce a classification of a hoax or not a Hoax based on the text inputted. Hoax news classification in this study uses five algorithms, namely Support Vector Machine, Naïve Bayes, Decision Tree, Logistic Regression, Stochastic Gradient Descent, and Neural Network (MLP). These five algorithms to produce the best algorithm that can use to detect hoax news, with the highest parameters, accuracy, F-measure, Precision, and recall. From the results of testing conducted on five classification algorithms produced shows that the NN-MPL algorithm has an average of 93% for the value of accuracy, F-Measure, and Precision, the highest compared to five other algorithms, but for the highest Recall value generated from the algorithm SVM which is 94%. the results of this experiment show that different effects for different classifiers, and that means that the more hoax data used as training data, the more accurate the system calculates accuracy in more detail.

Sign in / Sign up

Export Citation Format

Share Document