Although microscopy imaging is among the oldest techniques in the life sciences, its use as a quantifiable source of information has been limited to measure only certain specific phenotypes based on the biological question at hand. In recent years, several research groups have demonstrated that the data-rich information present in images can be used to create signatures of the state of a cell or other biological systems. Its applications can thus have a transformational impact on drug discovery as well as basic science. This field of image-based profiling is still in its infancy, and its adoption has been slow, primarily due to the data analysis challenges posed, the large imaging experiments required, and the lack of ground-truth data sets where relationships among images are known. Our goal is therefore to bring image-based profiling into the mainstream by organizing the community working on this technology.
We value interdisciplinary thinking and consider it critical to the success of this field. We strive to be a truly global society, preserving ethnic and geographical diversity and balance across demographic groups. By mandating that at least a fourth of the board comprise of early career researchers, we will encourage them to play a key role in decision-making. Individuals are considered early career if they either don’t hold a PhD, or if they do, then are within eight years of receiving the degree.
CytoDS will grant the use of its name and its full support to the organization of an annual scientific meeting of the image-based profiling research community. The purpose of these meetings is discussing recent advances in the field and fostering communication within the research community. Connections with other domains are forge by inviting experts from those fields to give a presentation at the symposium. The agenda is created by the organizers, while CytoDS will assist in fundraising, outreach and general organizational questions. The location of the meeting should be chosen based on an online poll sent to all members of the society, and must adhere to the equal rights policies of the society.
To push the limits of current and nurture the development of new cutting edge computational methods for analysis of image-based profiling experiments, CytoDS will assist in the organization of Hackathons. These can, but not have to, be connected to the annual conferences of the image-based profiling community. The content shall be decided by the organizers, but CytoDS will lend its name and assistance in fundraising, outreach and general organizational questions. The location of the Hackathon should vary each time and it must adhere to the equal rights policies of the society.
The visibility of CytoDS will be provided by a shared web-presence. The presence may be hosted in the public domain and shall be accessible freely for anybody. The web presence represents the society and all its members. It provides a framework for showcasing the societies activities and provides a framework for all other related web-hosted communication (e.g. forums).
Community forums and email lists
CytoDS aims to connect researchers worldwide in a vibrant community with many open opportunities to share and discuss ideas. Thus, in order to bridge geographical distances there should be a actively maintained web-based internet presence and community forum for scientific exchange. The forums and e-mail lists shall be moderated such that it is ensured everybody obeys to the values of CytoDS and the policies of good scientific practice. The forums can be hosted via public-corporate platforms or the societies own web-domain. Voluntary commitment of the societies members will ensure long-term sustained maintenance of the forum such that also its history does not get lost, when e.g. technology changes. We encourage using image.sc as discussion platform for image-based profiling.
Interconnecting academia and industry
To foster wide adoption and broad applications of results from research on image-based profiling methods, CytoDS aims to create open communication channels between academia and industrial partners. Researchers from both, industry as well as academia, are encouraged to contribute to discussions and participate in meetings.
Educating the next generation
While CytoDS aims to foster new technologies, it also sets out to actively pass down the current state of the art to the next generation of scientists.To this end, it actively seeks early career researchers to contribute to running the society. As well, members of the society are encouraged to reach out the next generation to raise awareness of the societies actions, meetings, and knowledge collections (e.g. web-resource collections). Hackathons and workshops during scientific meeting organized by CytoDS should further increase the accessibility of image-based profiling research.
Connections to other scientific communities
We aim to draw in other researchers from various different communities whose techniques / knowledge might be profitably applied to the field of image based profiling. This will ultimately lead to the mutual benefit of all communities and brings all research areas a step forward.
Machine learning and computer vision
Although some techniques from these fields, such as image filtering/preprocessing or feature extraction are readily applied to image-based profiling data, the high dimensional structure of image-based data demands that this community develop new techniques that exploit the multivariate nature of image information to gain new biological insights.
Other profiling modalities
Currently, the L1000 transcriptional profiling assay is the only viable technology in this space. Image-based profiling data presents this community the opportunity to add a complementary data source and methods. developed by the L1000 community to the field of image-based profiling.
The high-content screening community has several decades of experience in running large image-based experiments. Currently, nearly all image-based profiling datasets are likely generated by individuals who identify as part of this community. Their inputs on planning large-scale experiments, handling batch effects, and interfacing with other groups when executing a project, can provide invaluable inputs to inform the data analysis performed downstream.
Due to the enormous amount of data produced during an image based high-throughput experiment this community also builds on a shared knowledge base of how to handle, analyze and make sense of terabytes of data from single experiments, particularly at the single-cell level. This can be invaluable for other communities just starting with data production of such dimensions (e.g. single-cell sequencing and transcriptomics).
Society financial/legal structure
CytoData will not file for non-profit status and will not manage funds. Instead, each year, the symposium organizers will come up with a proposal and the committee will give permission to use the CytoData name.
The symposium strategy will be revisited each year.
Strategy for maintaining resources
CytoData will compile useful resources to help disseminate technical and scientific advances in the field. Details are provided in a dedicated section.