Srinivasan, K. | Search bioRxiv

bioRxiv2018

Prediction of Algal Blooms in the Great Lakes through a Convolution Neural Network of Remote Sensing Data

Harmful algal blooms (HABs) are the proliferation of algae due to eutrophication and have severe repercussions to the ecological balance in many water bodies, due to the toxins the algae produce. Additionally, the identification and prediction of these HABs has been a challenge in the scientific community due to the interactions between both biological and physical processes that cause the HABs. Here, we used remote sensing data to bypass these issues; remote sensing data provides significant information about the coverage of chlorophyll which can be used to locate HABs. Using this indicator of HABs, we trained a Convolution Neural Network (CNN) to identify nine types of algal blooms, using 25 epochs of 900 images, which can predict algal bloom shapes with an 80 percent accuracy. This approach of HAB identification can easily be applied to other aquatic ecosystems where remote sensing data is present.

ecology↗

bioRxiv2018

Deep Convolutional Network for Animal Sound Classification and Source Attribution using Dual Audio Recordings

We introduce an end-to-end feedforward convolutional neural network that is able to reliably classify the source and type of animal calls in a noisy environment using two streams of audio data after being trained on a dataset of modest size and imperfect labels. The data consists of audio recordings from captive marmoset monkeys housed in pairs, with several other cages nearby. Our network can classify both the call type and which animal made it with a single pass through a single network using raw spectrogram images as input. The network vastly increases data analysis capacity for researchers interested in studying marmoset vocalizations, and allows data collection in the home cage, in group housed animals.

animal behavior and cognition↗

bioRxiv2017

Exploring Single-Cell Data with Multitasking Deep Neural Networks

Biomedical researchers are generating high-throughput, high-dimensional single-cell data at a staggering rate. As costs of data generation decrease, experimental design is moving towards measurement of many different single-cell samples in the same dataset. These samples can correspond to different patients, conditions, or treatments. While scalability of methods to datasets of these sizes is a challenge on its own, dealing with large-scale experimental design presents a whole new set of problems, including batch effects and sample comparison issues. Currently, there are no computational tools that can both handle large amounts of data in a scalable manner (many cells) and at the same time deal with many samples (many patients or conditions). Moreover, data analysis currently involves the use of different tools that each operate on their own data representation, not guaranteeing a synchronized analysis pipeline. For instance, data visualization methods can be disjoint and mismatched with the clustering method. For this purpose, we present SAUCIE, a deep neural network that leverages the high degree of parallelization and scalability offered by neural networks, as well as the deep representation of data that can be learned by them to perform many single-cell data analysis tasks, all on a unified representation.\n\nA well-known limitation of neural networks is their interpretability. Our key contribution here are newly formulated regularizations (penalties) that render features learned in hidden layers of the neural network interpretable. When large multi-patient datasets are fed into SAUCIE, the various hidden layers contain denoised and batch-corrected data, a low dimensional visualization, unsupervised clustering, as well as other information that can be used to explore the data. We show this capability by analyzing a newly generated 180-sample dataset consisting of T cells from dengue patients in India, measured with mass cytometry. We show that SAUCIE, for the first time, can batch correct and process this 11-million cell data to identify cluster-based signatures of acute dengue infection and create a patient manifold, stratifying immune response to dengue on the basis of single-cell measurements.

bioinformatics↗