Using the antibody-antigen binding interface to train image-based deep neural networks for antibody-epitope classification

High-throughput B-cell sequencing has opened up new avenues for investigating complex mechanisms underlying our adaptive immune response. These technological advances drive data generation and the need to mine and analyze the information contained in these large datasets, in particular the identification of therapeutic antibodies (Abs) or those associated with disease exposure and protection. Here, we describe our efforts to use artificial intelligence (AI)-based image-analyses for prospective classification of Abs based solely on sequence information. We hypothesized that Abs recognizing the same part of an antigen share a limited set of features at the binding interface, and that the binding site regions of these Abs share share common structure and physicochemical property patterns that can serve as a “fingerprint” to recognize uncharacterized Abs. We combined large-scale sequence-based protein-structure predictions to generate ensembles of 3-D Ab models, reduced the Ab binding interface to a 2-D image (fingerprint), used pre-trained convolutional neural networks to extract features, and trained deep neural networks (DNNs) to classify Abs. We evaluated this approach using Ab sequences derived from human HIV and Ebola viral infections to differentiate between two Abs, Abs belonging to specific B-cell family lineages, and Abs with different epitope preferences. In addition, we explored a different type of DNN method to detect one class of Abs from a larger pool of Abs. Testing on Ab sets that had been kept aside during model training, we achieved average prediction accuracies ranging from 71–96% depending on the complexity of the classification task. The high level of accuracies reached during these classification tests suggests that the DNN models were able to learn a series of structural patterns shared by Abs belonging to the same class. The developed methodology provides a means to apply AI-based image recognition techniques to analyze high-throughput B-cell sequencing datasets (repertoires) for Ab classification.     

Using deep neural networks to model similarity between visual patterns: Application to fish sexual signals

The evolution of visual patterns is a frontier in the theory of sexual selection as we seek to understand the function of complex visual patterning in courtship. Recently, the sensory drive and sensory bias models of sexual selection have been applied to higher-level visual processing. One prediction of this application is that animals' sexual signals will mimic the visual statistics of their habitats. An enduring difficulty of testing predictions of visual pattern evolution is in developing quantitative methods for comparing patterns. Advances in artificial neural networks address this challenge by allowing for the direct comparison of images using both simple and complex features. Here, we use VGG19, an industry‑leading image classification network to test predictions of sensory drive, by comparing visual patterns in darter fish (Etheostoma spp.) to images of their habitats. We find that images of female darters are significantly more similar to images of their habitat than are images of males, supporting a role of camouflage in female patterning. We do not find direct evidence for sensory drive shaping the design of male patterns; however, this work demonstrates the utility of network methods for pattern analysis and suggests future directions for visual pattern research.     

Using ecological niche modelling to evaluate niche stability in deep time

Aim To investigate relative niche stability in species responses to various types of environmental pressure (biotic and abiotic) on geological time‐scales using the fossil record. Location The case study focuses on Late Ordovician articulate brachiopods of the Cincinnati Arch in eastern North America. Methods Species niches were modelled for a suite of fossil brachiopod species based on five environmental variables inferred from sedimentary parameters using GARP and Maxent . Niche stability was assessed by comparison of (1) the degree of overlap of species distribution models developed for a time‐slice and those generated by projecting niche models of the previous time‐slice onto environmental layers of a second time‐slice using GARP and Maxent , (2) Schoener’s D statistic, and (3) the similarity of the contribution of each environmental parameter within Maxent niche models between adjacent time‐slices. Results Late Ordovician brachiopod species conserved their niches with high fidelity during intervals of gradual environmental change but responded to inter‐basinal species invasions through niche evolution. Both native and invasive species exhibited similar levels of niche evolution in the invasion and post‐invasion intervals. Niche evolution was related mostly to decreased variance within the former ecological niche parameters rather than to shifts to new ecospace. Main conclusions Although the species examined exhibited morphological stasis during the study interval, high levels of niche conservatism were observed only during intervals of gradual environmental change. Rapid environmental change, notably inter‐basinal species invasions, resulted in high levels of niche evolution among the focal taxa. Both native and invasive species responded with similar levels of niche evolution during the invasion interval and subsequent environmental reorganization. The assumption of complete niche conservatism frequently employed in ecological niche modelling (ENM) analyses to forecast or hindcast species geographical distributions is more likely to be accurate for climate change studies than for invasive species analyses over geological time‐scales.     

Predicting enhancers with deep convolutional neural networks

Background

With the rapid development of deep sequencing techniques in the recent years, enhancers have been systematically identified in such projects as FANTOM and ENCODE, forming genome-wide landscapes in a series of human cell lines. Nevertheless, experimental approaches are still costly and time consuming for large scale identification of enhancers across a variety of tissues under different disease status, making computational identification of enhancers indispensable.

Results

To facilitate the identification of enhancers, we propose a computational framework, named DeepEnhancer, to distinguish enhancers from background genomic sequences. Our method purely relies on DNA sequences to predict enhancers in an end-to-end manner by using a deep convolutional neural network (CNN). We train our deep learning model on permissive enhancers and then adopt a transfer learning strategy to fine-tune the model on enhancers specific to a cell line. Results demonstrate the effectiveness and efficiency of our method in the classification of enhancers against random sequences, exhibiting advantages of deep learning over traditional sequence-based classifiers. We then construct a variety of neural networks with different architectures and show the usefulness of such techniques as max-pooling and batch normalization in our method. To gain the interpretability of our approach, we further visualize convolutional kernels as sequence logos and successfully identify similar motifs in the JASPAR database.

Conclusions

DeepEnhancer enables the identification of novel enhancers using only DNA sequences via a highly accurate deep learning model. The proposed computational framework can also be applied to similar problems, thereby prompting the use of machine learning methods in life sciences.

     

Single cortical neurons as deep artificial neural networks

1. Download : Download high-res image (228KB)
2. Download : Download full-size image

     

Control strategies in object manipulation tasks

The remarkable manipulative skill of the human hand is not the result of rapid sensorimotor processes, nor of fast or powerful effector mechanisms. Rather, the secret lies in the way manual tasks are organized and controlled by the nervous system. At the heart of this organization is prediction. Successful manipulation requires the ability both to predict the motor commands required to grasp, lift, and move objects and to predict the sensory events that arise as a consequence of these commands.     

Identifying wetland areas in historical maps using deep convolutional neural networks

The local environment and land usages have changed a lot during the past one hundred years. Historical documents and materials are crucial in understanding and following these changes. Historical documents are, therefore, an important piece in the understanding of the impact and consequences of land usage change. This, in turn, is important in the search of restoration projects that can be conducted to turn and reduce harmful and unsustainable effects originating from changes in the land-usage.This work extracts information on the historical location and geographical distribution of wetlands, from hand-drawn maps. This is achieved by using deep learning (DL), and more specifically a convolutional neural network (CNN). The CNN model is trained on a manually pre-labelled dataset on historical wetlands in the area of Jönköping county in Sweden. These are all extracted from the historical map called “Generalstabskartan”.The presented CNN performs well and achieves a F₁-score of 0.886 when evaluated using a 10-fold cross validation over the data. The trained models are additionally used to generate a GIS layer of the presumable historical geographical distribution of wetlands for the area that is depicted in the southern collection in Generalstabskartan, which covers the southern half of Sweden. This GIS layer is released as an open resource and can be freely used.To summarise, the presented results show that CNNs can be a useful tool in the extraction and digitalisation of non-textual information in historical documents, such as historical maps. A modern GIS material that can be used to further understand the past land-usage change is produced within this research. Previously, no material of this detail and extent have been available, due to the large effort needed to manually create such. However, with the presented resource better quantifications and estimations of historical wetlands that have been lost can be made.     

Assessment of locomotion in chlorine exposed mice by computer vision and neural networks

Assessment of locomotion following exposure of animals to noxious or painful stimuli can offer significant insights into underlying mechanisms of injury and the effectiveness of various treatments. We developed a novel method to track the movement of mice in two dimensions using computer vision and neural network algorithms. By using this system we demonstrated that mice exposed to chlorine (Cl(2)) gas developed impaired locomotion and increased immobility for up to 9 h postexposure. Postexposure administration of buprenorphine, a common analgesic agent, increased locomotion and decreased immobility times in Cl(2)- but not air-exposed mice, most likely by decreasing Cl(2)-induced pain. This method can be adapted to assess the effectiveness of various therapies following exposure to a variety of chemical and behavioral noxious stimuli.     

Motion vectors and deep neural networks for video camera traps

Commercial camera traps are usually triggered by a Passive Infra-Red (PIR) motion sensor necessitating a delay between triggering and the image being captured. This often seriously limits the ability to record images of small and fast moving animals. It also results in many “empty” images, e.g., owing to moving foliage against a background of different temperature. In this paper we detail a new triggering mechanism based solely on the camera sensor. This is intended for use by citizen scientists and for deployment on an affordable, compact, low-power Raspberry Pi computer (RPi). Our system introduces a video frame filtering pipeline consisting of movement and image-based processing. This makes use of Machine Learning (ML) feasible on a live camera stream on an RPi. We describe our free and open-source software implementation of the system; introduce a suitable ecology efficiency measure that mediates between specificity and recall; provide ground-truth for a video clip collection from camera traps; and evaluate the effectiveness of our system thoroughly. Overall, our video camera trap turns out to be robust and effective.     

Predicting enhancer-promoter interaction from genomic sequence with deep neural networks

Background: In the human genome, distal enhancers are involved in regulating target genes through proximal promoters by forming enhancer-promoter interactions. Although recently developed high-throughput experimental approaches have allowed us to recognize potential enhancer-promoter interactions genome-wide, it is still largely unclear to what extent the sequence-level information encoded in our genome help guide such interactions. Methods: Here we report a new computational method (named “SPEID”) using deep learning models to predict enhancer-promoter interactions based on sequence-based features only, when the locations of putative enhancers and promoters in a particular cell type are given. Results: Our results across six different cell types demonstrate that SPEID is effective in predicting enhancer-promoter interactions as compared to state-of-the-art methods that only use information from a single cell type. As a proof-of-principle, we also applied SPEID to identify somatic non-coding mutations in melanoma samples that may have reduced enhancer-promoter interactions in tumor genomes. Conclusions: This work demonstrates that deep learning models can help reveal that sequence-based features alone are sufficient to reliably predict enhancer-promoter interactions genome-wide.     

Leveraging high-throughput screening data,deep neural networks,and conditional generative adversarial networks to advance predictive toxicology

There are currently 85,000 chemicals registered with the Environmental Protection Agency (EPA) under the Toxic Substances Control Act, but only a small fraction have measured toxicological data. To address this gap, high-throughput screening (HTS) and computational methods are vital. As part of one such HTS effort, embryonic zebrafish were used to examine a suite of morphological and mortality endpoints at six concentrations from over 1,000 unique chemicals found in the ToxCast library (phase 1 and 2). We hypothesized that by using a conditional generative adversarial network (cGAN) or deep neural networks (DNN), and leveraging this large set of toxicity data we could efficiently predict toxic outcomes of untested chemicals. Utilizing a novel method in this space, we converted the 3D structural information into a weighted set of points while retaining all information about the structure. In vivo toxicity and chemical data were used to train two neural network generators. The first was a DNN (Go-ZT) while the second utilized cGAN architecture (GAN-ZT) to train generators to produce toxicity data. Our results showed that Go-ZT significantly outperformed the cGAN, support vector machine, random forest and multilayer perceptron models in cross-validation, and when tested against an external test dataset. By combining both Go-ZT and GAN-ZT, our consensus model improved the SE, SP, PPV, and Kappa, to 71.4%, 95.9%, 71.4% and 0.673, respectively, resulting in an area under the receiver operating characteristic (AUROC) of 0.837. Considering their potential use as prescreening tools, these models could provide in vivo toxicity predictions and insight into the hundreds of thousands of untested chemicals to prioritize compounds for HT testing.     

Unveiling functions of the visual cortex using task-specific deep neural networks

The human visual cortex enables visual perception through a cascade of hierarchical computations in cortical regions with distinct functionalities. Here, we introduce an AI-driven approach to discover the functional mapping of the visual cortex. We related human brain responses to scene images measured with functional MRI (fMRI) systematically to a diverse set of deep neural networks (DNNs) optimized to perform different scene perception tasks. We found a structured mapping between DNN tasks and brain regions along the ventral and dorsal visual streams. Low-level visual tasks mapped onto early brain regions, 3-dimensional scene perception tasks mapped onto the dorsal stream, and semantic tasks mapped onto the ventral stream. This mapping was of high fidelity, with more than 60% of the explainable variance in nine key regions being explained. Together, our results provide a novel functional mapping of the human visual cortex and demonstrate the power of the computational approach.     

Recurrent neural networks of integrate-and-fire cells simulating short-term memory and wrist movement tasks derived from continuous dynamic networks

Dynamic recurrent neural networks composed of units with continuous activation functions provide a powerful tool for simulating a wide range of behaviors, since the requisite interconnections can be readily derived by gradient descent methods. However, it is not clear whether more realistic integrate-and-fire cells with comparable connection weights would perform the same functions. We therefore investigated methods to convert dynamic recurrent neural networks of continuous units into networks with integrate-and-fire cells. The transforms were tested on two recurrent networks derived by backpropagation. The first simulates a short-term memory task with units that mimic neural activity observed in cortex of monkeys performing instructed delay tasks. The network utilizes recurrent connections to generate sustained activity that codes the remembered value of a transient cue. The second network simulates patterns of neural activity observed in monkeys performing a step-tracking task with flexion/extension wrist movements. This more complicated network provides a working model of the interactions between multiple spinal and supraspinal centers controlling motoneurons. Our conversion algorithm replaced each continuous unit with multiple integrate-and-fire cells that interact through delayed "synaptic potentials". Successful transformation depends on obtaining an appropriate fit between the activation function of the continuous units and the input-output relation of the spiking cells. This fit can be achieved by adapting the parameters of the synaptic potentials to replicate the input-output behavior of a standard sigmoidal activation function (shown for the short-term memory network). Alternatively, a customized activation function can be derived from the input-output relation of the spiking cells for a chosen set of parameters (demonstrated for the wrist flexion/extension network). In both cases the resulting networks of spiking cells exhibited activity that replicated the activity of corresponding continuous units. This confirms that the network solutions obtained through backpropagation apply to spiking networks and provides a useful method for deriving recurrent spiking networks performing a wide range of functions.     

Using weighted fixed neural networks for unsupervised fuzzy clustering

A novel algorithm for unsupervised fuzzy clustering is introduced. The algorithm uses a so-called Weighted Fixed Neural Network (WFNN) to store important and useful information about the topological relations in a given data set. The algorithm produces a weighted connected net, of weighted nodes connected by weighted edges, which reflects and preserves the topology of the input data set. The weights of the nodes and the edges in the resulting net are proportional to the local densities of data samples in input space. The connectedness of the net can be changed, and the higher the connectedness of the net is chosen, the fuzzier the system becomes. The new algorithm is computationally efficient when compared to other existing methods for clustering multi-dimensional data, such as color images.     

Haptic object perception: spatial dimensionality and relation to vision

Enabled by the remarkable dexterity of the human hand, specialized haptic exploration is a hallmark of object perception by touch. Haptic exploration normally takes place in a spatial world that is three-dimensional; nevertheless, stimuli of reduced spatial dimensionality are also used to display spatial information. This paper examines the consequences of full (three-dimensional) versus reduced (two-dimensional) spatial dimensionality for object processing by touch, particularly in comparison with vision. We begin with perceptual recognition of common human-made artefacts, then extend our discussion of spatial dimensionality in touch and vision to include faces, drawing from research on haptic recognition of facial identity and emotional expressions. Faces have often been characterized as constituting a specialized input for human perception. We find that contrary to vision, haptic processing of common objects is impaired by reduced spatial dimensionality, whereas haptic face processing is not. We interpret these results in terms of fundamental differences in object perception across the modalities, particularly the special role of manual exploration in extracting a three-dimensional structure.     

Global importance analysis: An interpretability method to quantify importance of genomic features in deep neural networks

Deep neural networks have demonstrated improved performance at predicting the sequence specificities of DNA- and RNA-binding proteins compared to previous methods that rely on k-mers and position weight matrices. To gain insights into why a DNN makes a given prediction, model interpretability methods, such as attribution methods, can be employed to identify motif-like representations along a given sequence. Because explanations are given on an individual sequence basis and can vary substantially across sequences, deducing generalizable trends across the dataset and quantifying their effect size remains a challenge. Here we introduce global importance analysis (GIA), a model interpretability method that quantifies the population-level effect size that putative patterns have on model predictions. GIA provides an avenue to quantitatively test hypotheses of putative patterns and their interactions with other patterns, as well as map out specific functions the network has learned. As a case study, we demonstrate the utility of GIA on the computational task of predicting RNA-protein interactions from sequence. We first introduce a convolutional network, we call ResidualBind, and benchmark its performance against previous methods on RNAcompete data. Using GIA, we then demonstrate that in addition to sequence motifs, ResidualBind learns a model that considers the number of motifs, their spacing, and sequence context, such as RNA secondary structure and GC-bias.