Synergy conformal prediction applied to large-scale bioactivity datasets and in federated learning.
Ontology highlight
ABSTRACT:
SUBMITTER: Norinder U
PROVIDER: S-EPMC8487527 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Machine learning methods may have the potential to significantly accelerate drug discovery. However, the increasing rate of new methodological approaches being published in the literature raises the fundamental question of how models should be benchmarked and validated. We reanalyze the data generated by a recently published large-scale comparison of machine learning models for bioactivity prediction and arrive at a somewhat different conclusion. We show that the performance of support vector machines is competitive with that of deep learning methods. Additionally, using a series of numerical experiments, we question the relevance of area under the receiver operating characteristic curve as a metric in virtual screening. We further suggest that area under the precision-recall curve should be used in conjunction with the receiver operating characteristic curve. Our numerical experiments also highlight challenges in estimating the uncertainty in model performance via scaffold-split nested cross validation.
| S-EPMC7292817 | biostudies-literature
Project description:The mobilization of large-scale datasets of specimen images and metadata through herbarium digitization provide a rich environment for the application and development of machine learning techniques. However, limited access to computational resources and uneven progress in digitization, especially for small herbaria, still present barriers to the wide adoption of these new technologies. Using deep learning to extract representations of herbarium specimens useful for a wide variety of applications, so-called "representation learning," could help remove these barriers. Despite its recent popularity for camera trap and natural world images, representation learning is not yet as popular for herbarium specimen images. We investigated the potential of representation learning with specimen images by building three neural networks using a publicly available dataset of over 2 million specimen images spanning multiple continents and institutions. We compared the extracted representations and tested their performance in application tasks relevant to research carried out with herbarium specimens. We found a triplet network, a type of neural network that learns distances between images, produced representations that transferred the best across all applications investigated. Our results demonstrate that it is possible to learn representations of specimen images useful in different applications, and we identify some further steps that we believe are necessary for representation learning to harness the rich information held in the worlds' herbaria.
| S-EPMC8794728 | biostudies-literature
Project description:Drug synergies are sought to identify combinations of drugs particularly beneficial. User-friendly software solutions that can assist analysis of large-scale datasets are required.CImbinator is a web-service that can aid in batch-wise and in-depth analyzes of data from small-scale and large-scale drug combination screens. CImbinator offers to quantify drug combination effects, using both the commonly employed median effect equation, as well as advanced experimental mathematical models describing dose response relationships.CImbinator is written in Ruby and R. It uses the R package drc for advanced drug response modeling. CImbinator is available at http://cimbinator.bioinfo.cnio.es , the source-code is open and available at https://github.com/Rbbt-Workflows/combination_index . A Docker image is also available at https://hub.docker.com/r/mikisvaz/rbbt-ci_mbinator/ .asmund.flobak@ntnu.no or miguel.vazquez@cnio.es.Supplementary data are available at Bioinformatics online.
| S-EPMC5860113 | biostudies-literature
Project description:There are growing concerns about the generalizability of machine learning classifiers in neuroimaging. In order to evaluate this aspect across relatively large heterogeneous populations, we investigated four disorders: Autism spectrum disorder (N = 988), Attention deficit hyperactivity disorder (N = 930), Post-traumatic stress disorder (N = 87) and Alzheimer's disease (N = 132). We applied 18 different machine learning classifiers (based on diverse principles) wherein the training/validation and the hold-out test data belonged to samples with the same diagnosis but differing in either the age range or the acquisition site. Our results indicate that overfitting can be a huge problem in heterogeneous datasets, especially with fewer samples, leading to inflated measures of accuracy that fail to generalize well to the general clinical population. Further, different classifiers tended to perform well on different datasets. In order to address this, we propose a consensus-classifier by combining the predictive power of all 18 classifiers. The consensus-classifier was less sensitive to unmatched training/validation and holdout test data. Finally, we combined feature importance scores obtained from all classifiers to infer the discriminative ability of connectivity features. The functional connectivity patterns thus identified were robust to the classification algorithm used, age and acquisition site differences, and had diagnostic predictive ability in addition to univariate statistically significant group differences between the groups. A MATLAB toolbox called Machine Learning in NeuroImaging (MALINI), which implements all the 18 different classifiers along with the consensus classifier is available from Lanka et al. (2019) The toolbox can also be found at the following URL: https://github.com/pradlanka/malini .
| S-EPMC7198352 | biostudies-literature
Project description:Structure-activity relationship modelling is frequently used in the early stage of drug discovery to assess the activity of a compound on one or several targets, and can also be used to assess the interaction of compounds with liability targets. QSAR models have been used for these and related applications over many years, with good success. Conformal prediction is a relatively new QSAR approach that provides information on the certainty of a prediction, and so helps in decision-making. However, it is not always clear how best to make use of this additional information. In this article, we describe a case study that directly compares conformal prediction with traditional QSAR methods for large-scale predictions of target-ligand binding. The ChEMBL database was used to extract a data set comprising data from 550 human protein targets with different bioactivity profiles. For each target, a QSAR model and a conformal predictor were trained and their results compared. The models were then evaluated on new data published since the original models were built to simulate a "real world" application. The comparative study highlights the similarities between the two techniques but also some differences that it is important to bear in mind when the methods are used in practical drug discovery applications.
| S-EPMC6690068 | biostudies-literature
Project description:Genomic data analysis across multiple cloud platforms is an ongoing challenge, especially when large amounts of data are involved. Here, we present Swarm, a framework for federated computation that promotes minimal data motion and facilitates crosstalk between genomic datasets stored on various cloud platforms. We demonstrate its utility via common inquiries of genomic variants across BigQuery in the Google Cloud Platform (GCP), Athena in the Amazon Web Services (AWS), Apache Presto and MySQL. Compared to single-cloud platforms, the Swarm framework significantly reduced computational costs, run-time delays and risks of security breach and privacy violation.
| S-EPMC8143397 | biostudies-literature
Project description:Earthquakes are a natural phenomenon which may cause significant loss of life and infrastructure. Researchers have applied multiple artificial intelligence based techniques to predict earthquakes, but high accuracies could not be achieved due to the huge size of multidimensional data, communication delays, transmission latency, limited processing capacity and data privacy issues. Federated learning (FL) is a machine learning (ML) technique that provides an opportunity to collect and process data onsite without compromising on data privacy and preventing data transmission to the central server. The federated concept of obtaining a global data model by aggregation of local data models inherently ensures data security, data privacy, and data heterogeneity. In this article, a novel earthquake prediction framework using FL has been proposed. The proposed FL framework has given better performance over already developed ML based earthquake predicting models in terms of efficiency, reliability, and precision. We have analyzed three different local datasets to generate multiple ML based local data models. These local data models have been aggregated to generate global data model on the central FL server using FedQuake algorithm. Meta classifier has been trained at the FL server on global data model to generate more accurate earthquake predictions. We have tested the proposed framework by analyzing multidimensional seismic data within 100 km radial area from 34.708° N, 72.5478° E in Western Himalayas. The results of the proposed framework have been validated against instrumentally recorded regional seismic data of last thirty-five years, and 88.87% prediction accuracy has been recorded. These results obtained by the proposed framework can serve as a useful component in the development of earthquake early warning systems.
| S-EPMC8176529 | biostudies-literature
Project description:The B-rapidly accelerated fibrosarcoma (BRAF) is a proto-oncogene that plays a vital role in cell signaling and growth regulation. Identifying a potent BRAF inhibitor can enhance therapeutic success in high-stage cancers, particularly metastatic melanoma. In this study, we proposed a stacking ensemble learning framework for the accurate prediction of BRAF inhibitors. We obtained 3857 curated molecules with BRAF inhibitory activity expressed as a predicted half-maximal inhibitory concentration value (pIC50) from the ChEMBL database. Twelve molecular fingerprints from PaDeL-Descriptor were calculated for model training. Three machine learning algorithms including extreme gradient boosting, support vector regression, and multilayer perceptron were utilized for constructing new predictive features (PFs). The meta-ensemble random forest regression, called StackBRAF, was created based on the 36 PFs. The StackBRAF model achieves lower mean absolute error (MAE) and higher coefficient of determination (R2 and Q2) than the individual baseline models. The stacking ensemble learning model provides good y-randomization results, indicating a strong correlation between molecular features and pIC50. An applicability domain of the model with an acceptable Tanimoto similarity score was also defined. Moreover, a large-scale high-throughput screening of 2123 FDA-approved drugs against the BRAF protein was successfully demonstrated using the StackBRAF algorithm. Thus, the StackBRAF model proved beneficial as a drug design algorithm for BRAF inhibitor drug discovery and drug development.
| S-EPMC10268632 | biostudies-literature
Project description:BackgroundMetagenomics is an approach to the characterization of microbial genomes via the direct isolation of genomic sequences from the environment without prior cultivation. The amount of metagenomic sequence data is growing fast while computational methods for metagenome analysis are still in their infancy. In contrast to genomic sequences of single species, which can usually be assembled and analyzed by many available methods, a large proportion of metagenome data remains as unassembled anonymous sequencing reads. One of the aims of all metagenomic sequencing projects is the identification of novel genes. Short length, for example, Sanger sequencing yields on average 700 bp fragments, and unknown phylogenetic origin of most fragments require approaches to gene prediction that are different from the currently available methods for genomes of single species. In particular, the large size of metagenomic samples requires fast and accurate methods with small numbers of false positive predictions.ResultsWe introduce a novel gene prediction algorithm for metagenomic fragments based on a two-stage machine learning approach. In the first stage, we use linear discriminants for monocodon usage, dicodon usage and translation initiation sites to extract features from DNA sequences. In the second stage, an artificial neural network combines these features with open reading frame length and fragment GC-content to compute the probability that this open reading frame encodes a protein. This probability is used for the classification and scoring of gene candidates. With large scale training, our method provides fast single fragment predictions with good sensitivity and specificity on artificially fragmented genomic DNA. Additionally, this method is able to predict translation initiation sites accurately and distinguishes complete from incomplete genes with high reliability.ConclusionLarge scale machine learning methods are well-suited for gene prediction in metagenomic DNA fragments. In particular, the combination of linear discriminants and neural networks is promising and should be considered for integration into metagenomic analysis pipelines. The data sets can be downloaded from the URL provided (see Availability and requirements section).
| S-EPMC2409338 | biostudies-literature
Project description:Understanding how congestion at one location can cause ripples throughout large-scale transportation network is vital for transportation researchers and practitioners to pinpoint traffic bottlenecks for congestion mitigation. Traditional studies rely on either mathematical equations or simulation techniques to model traffic congestion dynamics. However, most of the approaches have limitations, largely due to unrealistic assumptions and cumbersome parameter calibration process. With the development of Intelligent Transportation Systems (ITS) and Internet of Things (IoT), transportation data become more and more ubiquitous. This triggers a series of data-driven research to investigate transportation phenomena. Among them, deep learning theory is considered one of the most promising techniques to tackle tremendous high-dimensional data. This study attempts to extend deep learning theory into large-scale transportation network analysis. A deep Restricted Boltzmann Machine and Recurrent Neural Network architecture is utilized to model and predict traffic congestion evolution based on Global Positioning System (GPS) data from taxi. A numerical study in Ningbo, China is conducted to validate the effectiveness and efficiency of the proposed method. Results show that the prediction accuracy can achieve as high as 88% within less than 6 minutes when the model is implemented in a Graphic Processing Unit (GPU)-based parallel computing environment. The predicted congestion evolution patterns can be visualized temporally and spatially through a map-based platform to identify the vulnerable links for proactive congestion mitigation.
| S-EPMC4363621 | biostudies-literature