Dataset Information

Harnessing Deep Learning in Ecology: An Example Predicting Bark Beetle Outbreaks.

ABSTRACT: Addressing current global challenges such as biodiversity loss, global change, and increasing demands for ecosystem services requires improved ecological prediction. Recent increases in data availability, process understanding, and computing power are fostering quantitative approaches in ecology. However, flexible methodological frameworks are needed to utilize these developments towards improved ecological prediction. Deep learning is a rapidly evolving branch of machine learning, yet has received only little attention in ecology to date. It refers to the training of deep neural networks (DNNs), i.e. artificial neural networks consisting of many layers and a large number of neurons. We here provide a reproducible example (including code and data) of designing, training, and applying DNNs for ecological prediction. Using bark beetle outbreaks in conifer-dominated forests as an example, we show that DNNs are well able to predict both short-term infestation risk at the local scale and long-term outbreak dynamics at the landscape level. We furthermore highlight that DNNs have better overall performance than more conventional approaches to predicting bark beetle outbreak dynamics. We conclude that DNNs have high potential to form the backbone of a comprehensive disturbance forecasting system. More broadly, we argue for an increased utilization of the predictive power of DNNs for a wide range of ecological problems.

SUBMITTER: Rammer W

PROVIDER: S-EPMC6827389 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Harnessing Deep Learning in Ecology: An Example Predicting Bark Beetle Outbreaks.

Rammer Werner W Seidl Rupert R

Frontiers in plant science 20191028

Addressing current global challenges such as biodiversity loss, global change, and increasing demands for ecosystem services requires improved ecological prediction. Recent increases in data availability, process understanding, and computing power are fostering quantitative approaches in ecology. However, flexible methodological frameworks are needed to utilize these developments towards improved ecological prediction. Deep learning is a rapidly evolving branch of machine learning, yet has recei ...[more]

PMID: 31719829

Similar Datasets

Project description:Bark beetle outbreaks have intensified in many forests around the globe in recent years. Yet, the legacy of these disturbances for future forest development remains unclear. Bark beetle disturbances are expected to increase further because of climate change. Consequently, feedbacks within the disturbance regime are of growing interest, for example, whether bark beetle outbreaks are amplifying future bark beetle activity (through the initiation of an even-aged cohort of trees) or dampening it (through increased structural and compositional diversity).We studied bark beetle-vegetation-climate interactions in the Bavarian Forest National Park (Germany), an area characterised by unprecedented bark beetle activity in the recent past. We simulated the effect of future bark beetle outbreaks on forest structure and composition and analysed how disturbance-mediated forest dynamics influence future bark beetle activity under different scenarios of climate change. We used process-based simulation modelling in combination with machine learning to disentangle the long-term interactions between vegetation, climate and bark beetles at the landscape scale.Disturbances by the European spruce bark beetle were strongly amplified by climate change, increasing between 59% and 221% compared to reference climate. Bark beetle outbreaks reduced the dominance of Norway spruce (Picea abies (L.) Karst.) on the landscape, increasing compositional diversity. Disturbances decreased structural diversity within stands (α diversity) and increased structural diversity between stands (β diversity). Overall, disturbance-mediated changes in forest structure and composition dampened future disturbance activity (a reduction of up to -67%), but were not able to fully compensate for the amplifying effect of climate change. Synthesis. Our findings indicate that the recent disturbance episode at the Bavarian Forest National Park was caused by a convergence of highly susceptible forest structures with climatic conditions favourable for bark beetle outbreaks. While future climate is increasingly conducive to massive outbreaks, the emerging landscape structure is less and less likely to support them. This study improves our understanding of the long-term legacies of ongoing bark beetle disturbances in Central Europe. It indicates that increased diversity provides an important dampening feedback, and suggests that preventing disturbances or homogenizing post-disturbance forests could elevate the future susceptibility to large-scale bark beetle outbreaks.

Project description:Planning forest management relies on predicting insect outbreaks such as mountain pine beetle, particularly in the intermediate-term future, e.g., 5-year. Machine-learning algorithms are potential solutions to this challenging problem due to their many successes across a variety of prediction tasks. However, there are many subtle challenges in applying them: identifying the best learning models and the best subset of available covariates (including time lags) and properly evaluating the models to avoid misleading performance-measures. We systematically address these issues in predicting the chance of a mountain pine beetle outbreak in the Cypress Hills area and seek models with the best performance at predicting future 1-, 3-, 5- and 7-year infestations. We train nine machine-learning models, including two generalized boosted regression trees (GBM) that predict future 1- and 3-year infestations with 92% and 88% AUC, and two novel mixed models that predict future 5- and 7-year infestations with 86% and 84% AUC, respectively. We also consider forming the train and test datasets by splitting the original dataset randomly rather than using the appropriate year-based approach and show that this may obtain models that score high on the test dataset but low in practice, resulting in inaccurate performance evaluations. For example, a k-nearest neighbor model with the actual performance of 68% AUC, scores the misleadingly high 78% on a test dataset obtained from a random split, but the more accurate 66% on a year-based split. We then investigate how the prediction accuracy varies with respect to the provided history length of the covariates and find that neural network and naive Bayes, predict more accurately as history-length increases, particularly for future 1- and 3-year predictions, and roughly the same holds with GBM. Our approach is applicable to other invasive species. The resulting predictors can be used in planning forest and pest management and planning sampling locations in field studies.

Project description:BackgroundEpitope prediction is a useful approach in cancer immunology and immunotherapy. Many computational methods, including machine learning and network analysis, have been developed quickly for such purposes. However, regarding clinical applications, the existing tools are insufficient because few of the predicted binding molecules are immunogenic. Hence, to develop more potent and effective vaccines, it is important to understand binding and immunogenic potential. Here, we observed that the interactive association constituted by human leukocyte antigen (HLA)-peptide pairs can be regarded as a network in which each HLA and peptide is taken as a node. We speculated whether this network could detect the essential interactive propensities embedded in HLA-peptide pairs. Thus, we developed a network-based deep learning method called DeepNetBim by harnessing binding and immunogenic information to predict HLA-peptide interactions.ResultsQuantitative class I HLA-peptide binding data and qualitative immunogenic data (including data generated from T cell activation assays, major histocompatibility complex (MHC) binding assays and MHC ligand elution assays) were retrieved from the Immune Epitope Database database. The weighted HLA-peptide binding network and immunogenic network were integrated into a network-based deep learning algorithm constituted by a convolutional neural network and an attention mechanism. The results showed that the integration of network centrality metrics increased the power of both binding and immunogenicity predictions, while the new model significantly outperformed those that did not include network features and those with shuffled networks. Applied on benchmark and independent datasets, DeepNetBim achieved an AUC score of 93.74% in HLA-peptide binding prediction, outperforming 11 state-of-the-art relevant models. Furthermore, the performance enhancement of the combined model, which filtered out negative immunogenic predictions, was confirmed on neoantigen identification by an increase in both positive predictive value (PPV) and the proportion of neoantigen recognition.ConclusionsWe developed a network-based deep learning method called DeepNetBim as a pan-specific epitope prediction tool. It extracted the attributes of the network as new features from HLA-peptide binding and immunogenic models. We observed that not only did DeepNetBim binding model outperform other updated methods but the combination of our two models showed better performance. This indicates further applications in clinical practice.

Dataset Information

Harnessing Deep Learning in Ecology: An Example Predicting Bark Beetle Outbreaks.

Publications

Harnessing Deep Learning in Ecology: An Example Predicting Bark Beetle Outbreaks.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets