Collaborative Research: EAGER: Exploring beyond visualization: Data sonification of bacterial chemotaxis patterns

在这个大数据时代，前所未有的信息量正在以压倒研究人员以有意义的方式处理数据的能力的速度收集。在过去的三十年里，将数字流转换为图形表示已被证明是有用的，可以识别复杂数据集的趋势，如天气模式，股票市场波动和流感流行。虽然可视化是一种强大的数据分析方法，但并非所有数据都适合可视化。声化，即信息到声音的映射，是从视觉上混乱的数据中提取有用信息的另一种方法。数据声化的一个熟悉的例子是盖革计数器，它将不可见的伽马辐射转换为可听频率的滴答声。该项目展示了超声波在研究微生物如何游向对其生存至关重要的营养物质中的效用。其目标是促进更广泛地使用声化技术来分析生物研究界的大数据。数据的声音化还可以提高公众的科学素养和公众对科学技术的参与。正如朗朗上口的希格斯玻色子曲调所证明的那样，超声数据使亚原子粒子的发现更容易为公众所接受。在这个项目中，声音和音乐被用来提供一种媒介，通过这种媒介，以一种欢迎的方式让小学学龄儿童参与科学的兴奋。另一个值得注意的方面是，数据声化提供了一个方便的平台，使视力受损的个人参与研究。该项目汇集了生物系统工程和数字音乐创作方面的专业知识，为交叉培训的学生研究助理提供了多样化的视角。在该项目中，声化被用于检测暴露于化学刺激（即趋化性）后微生物游泳模式的变化。当通过显微镜观察一群游动的微生物时，它们的运动看起来很混乱，使得单个有机体的路径的细微变化在真实的时间内无法辨别。通过将视觉图像实时映射到频域，可以将混乱的视觉运动转换为听觉声音中的可辨别差异。具体的项目目标是：（1）确定在超声数据中检测到的细菌游动运动的特征;（2）优化视频显微镜设置和视频过滤器，以提高所收集数据的信噪比;（3）真实的时间超声数据，以便向观察者同时提供音频和视频输入;（4）评估针对具有不同游动行为的细菌的数据声化算法的鲁棒性;以及（5）筛选训练集以外的趋化性微生物，以评估声化过程的成功。这项工作的一个成果将是一个平台，以实时生成与视觉观察同步的超声数据，以允许高通量筛选各种物种对不同浓度范围内不同化学效应物的趋化反应。另一个，也许更有影响力的结果，该奖项由分子和细胞生物学部的系统和合成生物学群以及细胞和动力学群共同资助。该奖项反映了NSF的法定使命，并通过评估被认为值得支持使用基金会的知识价值和更广泛的影响审查标准。

In this Era of Big Data an unprecedented amount of information is being collected at rates that are overwhelming researchers' capacity to process data in meaningful ways. Converting streams of numbers into graphical representations has proven to be useful over the past three decades to identify trends in complex data sets such as weather patterns, stock market fluctuations, and flu epidemics. While visualization is a powerful approach to data analysis, not all data are amenable to visualization. Sonification, the mapping of information to sound, is an alternative method for extracting useful information from visually chaotic data. One familiar example of data sonification is a Geiger counter that converts invisible gamma radiation to an audible frequency of clicks. This project demonstrates the utility of sonification in a study of how microbes swim toward nutrients that are critical for their survival. The goal is to promote more widespread use of sonification to analyze big data within the biological research community. Sonification of data can also increase public scientific literacy and public engagement with science and technology. As demonstrated by the catchy Higgs Boson tune, sonified data made the discovery of subatomic particles more accessible to the public. Sound and music are used in this project to provide a medium through which to engage elementary school-age children in a welcoming manner about the excitement of science. Another notable aspect is that data sonification provides a convenient platform to engage sight-impaired individuals in research. The project brings together expertise in biological systems engineering and digital music composition that provide diverse perspectives for cross-training student research assistants.In this project sonification is used to detect changes in the swimming patterns of microorganisms upon exposure to a chemical stimulus (i.e. chemotaxis). When examining a population of swimming microbes through a microscope the movement appears chaotic, making subtle changes in the paths of individual organisms impossible to discern in real time. By mapping visual images to the frequency domain in real-time one can transform the chaotic visual motion to discernible differences in auditory sounds. The specific project objectives are to: (1) identify the features of bacterial swimming motion that are detected in sonified data; (2) optimize video microscopy settings and video filters to enhance the signal-to-noise ratio of the data collected; (3) sonify data in real time to allow simultaneous audio and visual input to an observer; (4) evaluate the robustness of data sonification algorithms for bacteria that have different swimming behaviors; and (5) screen microbes for chemotaxis beyond the training set to evaluate the success of the sonification process. One outcome of this work will be a platform to generate sonified data in real-time that is synchronous with visual observations to allow high-throughput screening of chemotactic responses for various species to different chemoeffectors over a range of concentrations. Another, and perhaps more impactful outcome, will be to significantly expand the tools that biological scientists have at their disposal to identify patterns in complex data that they collect.This award is jointly funded by the Systems and Synthetic Biology Cluster and the Cellular and Dynamics Cluster in the Division of Molecular and Cellular Biology.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.