Low-dimensional learned feature spaces quantify individual and group differences in vocal repertoires

  1. Jack Goffinet
  2. Samuel Brudner
  3. Richard Mooney
  4. John Pearson  Is a corresponding author
  1. Duke University, United States

Abstract

Increases in the scale and complexity of behavioral data pose an increasing challenge for data analysis. A common strategy involves replacing entire behaviors with small numbers of handpicked, domain-specific features, but this approach suffers from several crucial limitations. For example, handpicked features may miss important dimensions of variability, and correlations among them complicate statistical testing. Here, by contrast, we apply the variational autoencoder (VAE), an unsupervised learning method, to learn features directly from data and quantify the vocal behavior of two model species: the laboratory mouse and the zebra finch. The VAE converges on a parsimonious representation that outperforms handpicked features on a variety of common analysis tasks, enables the measurement of moment-by-moment vocal variability on the timescale of tens of milliseconds in the zebra finch, provides strong evidence that mouse ultrasonic vocalizations do not cluster as is commonly believed, and captures the similarity of tutor and pupil birdsong with qualitatively higher fidelity than previous approaches. In all, we demonstrate the utility of modern unsupervised learning approaches to the quantification of complex and high-dimensional vocal behavior.

Data availability

Dataset 1---------Online, publicly available MUPET dataset: ~5GB Available at: https://github.com/mvansegbroeck/mupet/wiki/MUPET-wiki Figs: 2, 3, 4d-eDataset 2----------Single zebra finch data: ~200-400 MB of audio generated as part of work in progress in Mooney Lab. Figs: 2e-f, 4a-c, 5a, 5d, 6b-eDataset 3---------Mouse USV dataset: ~30-40 GB of audio generated as part of work in progress in Mooney Lab. Figs: 4fDataset 5---------This is a subset of dataset 3, taken from a single mouse: ~1GB of audio. Figs: 5b-e, 6aDataset 6---------10 zebra finch pupil/tutor pairs: ~60 GB of audio generated as part of work in progress in Mooney Lab. Figs: 7Upon acceptance, all Datasets 2-6 will be archived in the Duke Digital Repository (https://research.repository.duke.edu). DOI in process.

The following previously published data sets were used

Article and author information

Author details

  1. Jack Goffinet

    Computer Science, Duke University, Durham, United States
    Competing interests
    The authors declare that no competing interests exist.
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-6729-0848
  2. Samuel Brudner

    Neurobiology, Duke University, Durham, United States
    Competing interests
    The authors declare that no competing interests exist.
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-6043-9328
  3. Richard Mooney

    Department of Neurobiology, Duke University, Durham, United States
    Competing interests
    The authors declare that no competing interests exist.
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-3308-1367
  4. John Pearson

    Biostatistics & Bioinformatics, Neurobiology, Center for Cognitive Neuroscience, Psychology and Neuroscience, Electrical and Computer Engineering, Duke University, Durham, United States
    For correspondence
    john.pearson@duke.edu
    Competing interests
    The authors declare that no competing interests exist.
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-9876-7837

Funding

National Institute of Mental Health (R01-MH117778)

  • Richard Mooney

National Institute of Neurological Disorders and Stroke (R01-NS118424)

  • Richard Mooney
  • John Pearson

National Institute on Deafness and Other Communication Disorders (R01-DC013826)

  • Richard Mooney
  • John Pearson

National Institute of Neurological Disorders and Stroke (R01-NS099288)

  • Richard Mooney

Eunice Kennedy Shriver National Institute of Child Health and Human Development (F31-HD098772)

  • Samuel Brudner

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Ethics

Animal experimentation: All data generated in conjunction for this study were generated by experiments performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All of the animals were handled according to approved institutional animal care and use committee (IACUC) protocols of Duke University, protocol numbers A171-20-08 and A172-20-08.

Copyright

© 2021, Goffinet et al.

This article is distributed under the terms of the Creative Commons Attribution License permitting unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 4,937
    views
  • 571
    downloads
  • 72
    citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.

Download links

Share this article

https://doi.org/10.7554/eLife.67855

Further reading

    1. Computational and Systems Biology
    2. Genetics and Genomics
    Fangluo Chen, Dylan C Sarver ... G William Wong
    Research Article

    Obesity is a major risk factor for type 2 diabetes, dyslipidemia, cardiovascular disease, and hypertension. Intriguingly, there is a subset of metabolically healthy obese (MHO) individuals who are seemingly able to maintain a healthy metabolic profile free of metabolic syndrome. The molecular underpinnings of MHO, however, are not well understood. Here, we report that CTRP10/C1QL2-deficient mice represent a unique female model of MHO. CTRP10 modulates weight gain in a striking and sexually dimorphic manner. Female, but not male, mice lacking CTRP10 develop obesity with age on a low-fat diet while maintaining an otherwise healthy metabolic profile. When fed an obesogenic diet, female Ctrp10 knockout (KO) mice show rapid weight gain. Despite pronounced obesity, Ctrp10 KO female mice do not develop steatosis, dyslipidemia, glucose intolerance, insulin resistance, oxidative stress, or low-grade inflammation. Obesity is largely uncoupled from metabolic dysregulation in female KO mice. Multi-tissue transcriptomic analyses highlighted gene expression changes and pathways associated with insulin-sensitive obesity. Transcriptional correlation of the differentially expressed gene (DEG) orthologs in humans also shows sex differences in gene connectivity within and across metabolic tissues, underscoring the conserved sex-dependent function of CTRP10. Collectively, our findings suggest that CTRP10 negatively regulates body weight in females, and that loss of CTRP10 results in benign obesity with largely preserved insulin sensitivity and metabolic health. This female MHO mouse model is valuable for understanding sex-biased mechanisms that uncouple obesity from metabolic dysfunction.

    1. Computational and Systems Biology
    Huiyong Cheng, Dawson Miller ... Qiuying Chen
    Research Article

    Mass spectrometry imaging (MSI) is a powerful technology used to define the spatial distribution and relative abundance of metabolites across tissue cryosections. While software packages exist for pixel-by-pixel individual metabolite and limited target pairs of ratio imaging, the research community lacks an easy computing and application tool that images any metabolite abundance ratio pairs. Importantly, recognition of correlated metabolite pairs may contribute to the discovery of unanticipated molecules in shared metabolic pathways. Here, we describe the development and implementation of an untargeted R package workflow for pixel-by-pixel ratio imaging of all metabolites detected in an MSI experiment. Considering untargeted MSI studies of murine brain and embryogenesis, we demonstrate that ratio imaging minimizes systematic data variation introduced by sample handling, markedly enhances spatial image contrast, and reveals previously unrecognized metabotype-distinct tissue regions. Furthermore, ratio imaging facilitates identification of novel regional biomarkers and provides anatomical information regarding spatial distribution of metabolite-linked biochemical pathways. The algorithm described herein is applicable to any MSI dataset containing spatial information for metabolites, peptides or proteins, offering a potent hypothesis generation tool to enhance knowledge obtained from current spatial metabolite profiling technologies.