1. Cell Biology
  2. Computational and Systems Biology
Download icon

Active machine learning-driven experimentation to determine compound effects on protein patterns

  1. Armaghan W Naik
  2. Joshua D Kangas
  3. Devin P Sullivan
  4. Robert F Murphy  Is a corresponding author
  1. Carnegie Mellon University, United States
Research Article
  • Cited 23
  • Views 6,336
  • Annotations
Cite this article as: eLife 2016;5:e10047 doi: 10.7554/eLife.10047

Abstract

High throughput screening determines the effects of many conditions on a given biological target. Currently, to estimate the effects of those conditions on other targets requires either strong modeling assumptions (e.g. similarities among targets) or separate screens. Ideally, data-driven experimentation could be used to learn accurate models for many conditions and targets without doing all possible experiments. We have previously described an active machine learning algorithm that can iteratively choose small sets of experiments to learn models of multiple effects. We now show that, with no prior knowledge and with liquid handling robotics and automated microscopy under its control, this learner accurately learned the effects of 48 chemical compounds on the subcellular localization of 48 proteins while performing only 29% of all possible experiments. The results represent the first practical demonstration of the utility of active learning-driven biological experimentation in which the set of possible phenotypes is unknown in advance.

Article and author information

Author details

  1. Armaghan W Naik

    Computational Biology Department, Center for Bioimage Informatics, Carnegie Mellon University, Pittsburgh, United States
    Competing interests
    The authors declare that no competing interests exist.
  2. Joshua D Kangas

    Computational Biology Department, Center for Bioimage Informatics, Carnegie Mellon University, Pittsburgh, United States
    Competing interests
    The authors declare that no competing interests exist.
  3. Devin P Sullivan

    Computational Biology Department, Center for Bioimage Informatics, Carnegie Mellon University, Pittsburgh, United States
    Competing interests
    The authors declare that no competing interests exist.
  4. Robert F Murphy

    Computational Biology Department, Center for Bioimage Informatics, Carnegie Mellon University, Pittsburgh, United States
    For correspondence
    murphy@cmu.edu
    Competing interests
    The authors declare that no competing interests exist.

Reviewing Editor

  1. Uwe Ohler, Duke, Germany

Publication history

  1. Received: July 13, 2015
  2. Accepted: January 28, 2016
  3. Accepted Manuscript published: February 3, 2016 (version 1)
  4. Version of Record published: March 7, 2016 (version 2)

Copyright

© 2016, Naik 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

  • 6,336
    Page views
  • 1,141
    Downloads
  • 23
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, Scopus, PubMed Central.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)

Further reading

    1. Cell Biology
    Nilay Nandi et al.
    Research Advance Updated

    Neuronal health depends on quality control functions of autophagy, but mechanisms regulating neuronal autophagy are poorly understood. Previously, we showed that in Drosophila starvation-independent quality control autophagy is regulated by acinus (acn) and the Cdk5-dependent phosphorylation of its serine437 (Nandi et al., 2017). Here, we identify the phosphatase that counterbalances this activity and provides for the dynamic nature of acinus-serine437 (acn-S437) phosphorylation. A genetic screen identified six phosphatases that genetically interacted with an acn gain-of-function model. Among these, loss of function of only one, the PPM-type phosphatase Nil (CG6036), enhanced pS437-acn levels. Cdk5-dependent phosphorylation of acn-S437 in nil1 animals elevates neuronal autophagy and reduces the accumulation of polyQ proteins in a Drosophila Huntington’s disease model. Consistent with previous findings that Cd2+ inhibits PPM-type phosphatases, Cd2+ exposure elevated acn-S437 phosphorylation which was necessary for increased neuronal autophagy and protection against Cd2+-induced cytotoxicity. Together, our data establish the acn-S437 phosphoswitch as critical integrator of multiple stress signals regulating neuronal autophagy.

    1. Cell Biology
    2. Immunology and Inflammation
    Bochuan Li et al.
    Research Article Updated

    Atherosclerosis preferentially occurs in atheroprone vasculature where human umbilical vein endothelial cells are exposed to disturbed flow. Disturbed flow is associated with vascular inflammation and focal distribution. Recent studies have revealed the involvement of epigenetic regulation in atherosclerosis progression. N6-methyladenosine (m6A) is the most prevalent internal modification of eukaryotic mRNA, but its function in endothelial atherogenic progression remains unclear. Here, we show that m6A mediates the epidermal growth factor receptor (EGFR) signaling pathway during EC activation to regulate the atherosclerotic process. Oscillatory stress (OS) reduced the expression of methyltransferase like 3 (METTL3), the primary m6A methyltransferase. Through m6A sequencing and functional studies, we determined that m6A mediates the mRNA decay of the vascular pathophysiology gene EGFR which leads to EC dysfunction. m6A modification of the EGFR 3’ untranslated regions (3’UTR) accelerated its mRNA degradation. Double mutation of the EGFR 3’UTR abolished METTL3-induced luciferase activity. Adenovirus-mediated METTL3 overexpression significantly reduced EGFR activation and endothelial dysfunction in the presence of OS. Furthermore, thrombospondin-1 (TSP-1), an EGFR ligand, was specifically expressed in atheroprone regions without being affected by METTL3. Inhibition of the TSP-1/EGFR axis by using shRNA and AG1478 significantly ameliorated atherogenesis. Overall, our study revealed that METTL3 alleviates endothelial atherogenic progression through m6A-dependent stabilization of EGFR mRNA, highlighting the important role of RNA transcriptomics in atherosclerosis regulation.