Elsevier

Ecological Informatics

Volume 1, Issue 3, November 2006, Pages 219-228
Ecological Informatics

Pattern detection of movement behaviors in genotype variation of Drosophila melanogaster by using self-organizing map

https://doi.org/10.1016/j.ecoinf.2005.12.002Get rights and content

Abstract

The two dimensional movement tracks of STAT92E06346 mutant and two control strains (Oregon red (OR) and TM3) of Drosophila melonogaster were continuously observed with image processors. Subsequently Self-Organizing Map (SOM) was implemented to patterning of responding behaviors of the tested specimens. Movement behaviors were accordingly revealed in different strains and sex. SOM showed difference in degree of grouping in behaviors in different genotypes. Visualization through SOM further characterized the clusters of specimens with the variables regarding activities and spatial information. The study demonstrated that techniques in data mining in artificial neural networks could be a useful tool for analyzing complex behaviors induced by changes in genetic information.

Introduction

Drosophila has been a suitable system to answer behavioral questions based on the availability of mutant phenotypes (Meehan and Wilson, 1987). Rapid development of molecular genetics has led to the identification of a large number of genes and contributed greatly to revealing behavioral changes corresponding to genotype variation (Pflugfelder, 1998, Ganguly et al., 2003). Especially, locomotion of Drosophila has been an efficient target for studying the gene-behavior relationships in various aspects encompassing such areas as circadian rhythm (Park et al., 2000, Williarns and Seghal, 2001), courtship (Yamamoto and Nakano, 1999, Greenspan and Feveur, 2000), and (Kaneuchi et al., 2001) learning and memory (Dubnau and Tully, 1998, Waddell and Quinn, 2001).

Behavioral data, however, are in general complex and difficult to analyze. In this study, we demonstrate the use of artificial neural networks in extracting information from complex movement data to provide comprehensive feature on behaviors in different genotypes (De Belle and Sokolowski, 1987). We used the specific gene, JAK/STAT (Janus Kinase/Signal Transducer and Activator of Transcription), which plays critical roles in the mammalian immune system and the development of signaling pathways (Akira, 1999, Catlett -Falcone et al., 1999, Imada and Leonard, 2000). The STAT proteins are activated through phosphorylation on a tyrosine residue by JAKs. JAK kinase, encoded by the gene hopscotch (Binari and Perrimon, 1994), and a STAT protein, encoded by the gene marelle also known as Dstat or Stat92E (Hou et al., 1996, Yan et al., 1996), have been characterized in Drosophila. Activation of the JAK/STAT pathway has been observed in response to generation of intracellular reactive oxygen species (ROS) (Simon et al., 1998, Yang, in press) and exogenous hydrogen peroxide (H2O2) (Carballo et al., 1999). ROS are generated intracellularly through a variety of processes as by-products of normal aerobic metabolism or as second messengers in various signal transduction pathways (Martindale and Holbrook, 2002) in the pathogenesis of many human diseases such as acute respiratory distress syndrome, Parkinson's diseases, pulmonary fibrosis, and Alzheimer's disease. Especially ROS are involved in insulin and leptin signaling pathways (Carvalheira et al., 2003, Rosenblum et al., 1996, Storz et al., 1999). Leptin is an adipocyte-derived hormone that plays a key role in the control of food intake and energy expenditure (Halaas et al., 1995, Flier, 1998).

Considering its association with aerobic metabolism and signal transduction by ROS and in leptin signaling pathways, STAT could be closely related to the expression of various behaviors of Drosophila including changes in activity. In this study, we explored the effect of the gene in STAT on movement behaviors of Drosophila in semi-natural conditions of observation chambers. Recently numerous accounts of studying the behavior of animals through computer observation techniques have been reported (e. g., Kaneuchi et al., 2003, Kwak et al., 2002); however, the molecular and genetic basis of behavior has remained elusive so far.

In this study, we implemented SOM to classify the movement patterns corresponding to different genotypes. Through unsupervised learning, SOM is effective in extracting features out of complex data sets and produces comprehensible low-dimensional maps for input data (Zurada, 1992). SOM has been widely used as of late in such areas as classification and ordination of ecological data (Chon et al., 1996, Chon et al., 2000, Park et al., 2003), recognition of speech (Kohonen et al., 1984, Kohonen et al., 1987), sentence understanding (Samarabandu and Jakubowicz, 1990), classification of sea-ice (Orlando et al., 1990) and even grouping of insect courtship songs (Neumann et al., 1990). Recently, the application of SOM has expanded to the molecular sciences, the analysis of gene expression in DNA microarray (Nikkila et al., 2002, Toronen et al., 1999), the classification of mass spectra data (Lohninger and Stanel, 1992), the application of drug discovery (Manallack and Livingstone, 1999) and the analysis of codon usage patterns of bacterial genomes (Wang et al., 2001). In this study, we explored how different genotypes were projected onto two dimensional movement data to elucidate gene effect, and demonstrated that SOM is feasible in patterning and visualizing complex behavioral data.

Section snippets

Test specimens

STAT92E06346 flies (STAT), mutant strain of Drosophila melanogaster, indicate the specimens with STAT heterozygotes (Hou et al., 1996). The precise STAT92E06346 fly genotype is ry506 P{ry+t7.2 = PZ} Stat92E06346/TM3, ryRK Sb1 Ser1 (provided from Bloomiton Stock Center). P{pz} is inserted into an intron that separates the promoter and first exon from the main body of the coding sequence (Hou et al., 1996). Homozygotic individuals die at the end of embryogenesis or as young larvae (Luo et al., 1997

Analysis of variables

As an initial step of study we analyzed the six variables covering activities and spatial information of the specimens. Fig. 2 shows differences in the variables in different strains and sex. The STAT heterozygote flies exhibited a significantly decreased activity compared with the control group (OR, TM3) (Fig. 2). The duration of movement (Fig. 2a) and the speed in the total duration of observation (Fig. 2b) were distinctively lower in STAT compared with the control group, OR and TM3, while

Discussion

In this study, we demonstrated characterization of behavioral data was possible with changes in the mutant genotypes of Drosophila. The movement patterns were accordingly revealed in different variables of activity and spatial location in different genotypes (OR, TM3, and STAT) (Fig. 2, Fig. 3, Fig. 4, Fig. 5, Fig. 6, Fig. 7). SOM was suitable for mining the complex behavioral data. Although specific matching between behaviors and genotypes was not clearly observed, different degrees of

Acknowledgment

This work was supported by Pusan National Research Grant.

References (61)

  • Y.-S. Park et al.

    Computational characterization of behavioral response of medaka (Oryzias latipes) treated with diazinon

    Aquat. Toxicol.

    (2005)
  • G.O. Pflugfelder

    Genetic lesions in Drosophila behavioural mutants

    Behav. Brain Res.

    (1998)
  • C. Schal et al.

    Search strategy of sex pheromone-stimulated male German cockroaches

    J. Insect Physiol.

    (1983)
  • P. Storz et al.

    Insulin selectively activates STAT5b, but not STAT5a, via a JAK2-independent signalling pathway in Kym-1 rhabdomyosarcoma cells

    Fed. Eur. Biochem. Soc. Lett. (FEBS Lett.)

    (1999)
  • P. Toronen et al.

    Analysis of gene expression data using self-organizing maps

    Fed. Eur. Biochem. Soc. Lett. (FEBS Lett.)

    (1999)
  • M.K. Tourtellot et al.

    The problem of move length and turn definition in analysis of orientation data

    J. Theor. Biol.

    (1991)
  • R. Yan et al.

    Identification of a Stat gene that functions in Drosophila development

    Cell

    (1996)
  • S. Akira

    Functional roles of STAT family proteins: lessons from knockout mice

    Stem Cells

    (1999)
  • W. Alt

    Correlation analysis of two-dimensional locomotion paths

  • Y.H. Belgacem et al.

    Disruption of insulin pathways alters trehalose level and abolishes sexual dimorphism in locomotor activity in Drosophila

    J. Neurobiol.

    (2006)
  • R. Binari et al.

    Stripe-specific regulation of pair-rule genes by hopscotch, aputative Jak family tyrosine kinase in Drosophila

    Genes Dev.

    (1994)
  • J.B. Carvalheira et al.

    Interaction between leptin and insulin signaling pathways differentially affects JAK-STAT and PI 3-kinase-mediated signaling in rat liver

    Biol. Chem. Jpn.

    (2003)
  • R. Catlett -Falcone et al.

    STAT proteins as novel targets for cancer therapy. Signal transducer an activator of transcription

    Curr. Opin. Oncol.

    (1999)
  • T.-S. Chon et al.

    Use of an artificial neural network to predict population dynamics of the forest-pest pine needle Gall Midge (Diptera: Cecidomyiida)

    Popul. Econ.

    (2000)
  • T.-S. Chon et al.

    Implementation of computational methods to pattern recognition of movement behavior of Blattlella germanica (Blattaria: Blattidae) treated with Ca2+ signal inducing chemicals

    Appl. Entomol. Zool.

    (2004)
  • R.D. Collins et al.

    Innate control of local search behaviour in the house fly, Musca domestica

    Physiol. Entomol.

    (1994)
  • J.S. De Belle et al.

    Heredity of rover/sitter: alternative foraging strategies of Drosophila melanogaster larvae

    Heredity

    (1987)
  • J. Dubnau et al.

    Gene discovery in Drosophila: new insights for learning and memory

    Annu. Rev. Neurosci.

    (1998)
  • J.S. Flier

    Clinical review 94: what's in a name? In search of leptin's physiologic role

    J. Clin. Endocrinol. Metab.

    (1998)
  • Cited by (7)

    • Application of temporal self-organizing maps to patterning short-time series of fish behavior responding to environmental stress

      2020, Ecological Modelling
      Citation Excerpt :

      Since Kohonen (1982) proposed, the SOM networks have been extensively applied in various fields (Park et al., 2003; De et al., 2014; López et al., 2012; Strickert et al., 2005; Tokutaka et al., 2015), including ecological science (Kalteh et al., 2008; Céréghino and Park, 2009; Chon, 2011). The movement of animals were specifically analyzed under chemical stress in association with SOM on different taxa, including fishes (e.g. Zebrafish, Medaka) (Lee et al., 2015; Park et al., 2005; Uehara et al., 2015; Tsai et al., 2017), insects (e.g. Drosophila, Termite) (Choi et al., 2006; Lee and Chon, 2011), benthic macroinvertebrates (Park et al., 2004; Son et al., 2006), and birds (Lee et al., 2006). In order to demonstrate visualization of intoxication or recovering states of organisms, SOMs were further applied to patterning SBRM data to cluster response behaviors according to chemical concentrations and measurement times (Liu et al., 2011; Wang et al., 2015).

    • Modelling animal behavior to monitor effects of stressors

      2015, Developments in Environmental Modelling
      Citation Excerpt :

      Sensitive species can be specifically selected to detect stressors with specific molecular characteristics. This is a new field that will likely advance with the development of molecular genetics (e.g., Choi et al., 2006; Liu et al., 2011a; Eom et al., 2014). On the other hand, some species could be used as general indicators to serve as an early warning system in cases in which stressors are mixed under field conditions (Gerhardt et al., 2003; Ren et al., 2007).

    • Self-Organizing Maps applied to ecological sciences

      2011, Ecological Informatics
      Citation Excerpt :

      At the organism level, SOMs have primarily been used to analyze behavioral data in response to the existence of environmental stressors (e.g., toxic chemicals). As described previously, SOMs has been used in the analysis of behavioral data in response to different genotypes (Choi et al., 2006). However, the majority of applications of SOMs to behavioral data have focused on the detection of response behaviors of individual animals in ecological risk assessment.

    View all citing articles on Scopus
    View full text