-
Notifications
You must be signed in to change notification settings - Fork 0
Code of Conduct
The goal of this community is to advance circuit neuroscience in the mosquito by collaboratively producing and analyzing the connectome of an adult female non-blood fed Aedes aegypti brain.
The Female Adult Aedes aegypti Brain (FAAB) dataset was conceptualized in 2018 by Meg Younger (BU), David Hildebrand (Rockefeller>UHouston), and Wei-Chung Allen Lee (BCH/Harvard). Data acquisition and alignment were completed in 2024 with technical support from Mohd Freezly. From 2022 to 2024, the Lee and Younger labs contracted Zetta AI to produce an automated segmentation of all neurons and to generate synapse predictions.
Early in 2025, Greg Jefferis (MRC LMB/Cambridge University) and Elizabeth Marin (Cambridge University) secured funding with Meg and Wei for collaborative reconstruction, annotation, and analysis from the Wellcome Trust, and their group became major contributors to proofreading and data management efforts. The primary goals of this £5M grant are to:
- produce a complete, high-quality connectome for an entire Aedes aegypti central brain through proofreading, expert validation, and annotation of the automated segmentation,
- obtain novel scientific insights into neural circuits underlying host-seeking behaviour as a model for multisensory integration, and
- perform comparative connectomics with extant Drosophila melanogaster brain volumes to identify conserved vs divergent cells and circuits over >160 my of evolution.
Achieving these ambitious goals will require large-scale, coordinated, collaborative proofreading and annotation efforts from all users of this dataset. Systematic proofreading and annotation work, with contributions from the four co-PIs and their teams, began with the start of the grant in May 2025.
The FAAB Community Steering Committee was formed to coordinate proofreading and annotation efforts. This Steering Committee consists of Wei-Chung Allen Lee <Wei-Chung_Lee@hms.harvard.edu>, Meg Younger <myounger@bu.edu>, David Hildebrand <david@hildebrand.name>, Greg Jefferis <jefferis@mrc-lmb.cam.ac.uk>, and Elizabeth Marin <em711@cam.ac.uk>.
Collaboratively producing and analyzing a connectome from the FAAB dataset depends on the work of many individuals, requiring mutual respect and trust to maintain a productive and collegial community.
It is critical that proofreading and annotation edits are made responsibly. Edits will be reviewed periodically and additional training may be required if an individual makes frequent errors. If an individual's edits cause ongoing problems for the community, access privileges may be revoked.
Open communication is a core tenet of this community. The Slack workspace is our primary method for communication. All active members of the community must check Slack regularly and have notifications turned on for direct messages and being tagged in posts.
Individuals are not permitted to use FAAB community data for any purpose without the explicit consent of the data owner(s) (see the Data ownership). Results of proofreading and annotation are immediately public to the community, similar to working in an open lab notebook, and should be considered unpublished work. Information derived from this work should not be shared with anyone who has not been authorized to have access to these data and who has not agreed to these terms. Breaking this rule is a breach of scientific ethics, and will lead to access privileges being immediately revoked.
We warmly welcome additional users to participate in proofreading and/or annotation, with the understanding that pursuing research interests in specific neurons and circuits should contribute towards the larger goal of a complete, high-quality connectome available to the worldwide research community. We intend to publish an initial dataset paper describing the partially proofread connectome, while a later paper will describe the fully proofread and annotated connectome. Co-authorship of these papers will be evaluated based on the level of contribution, consistent with scientific standards.
As a first step toward joining the FAAB community, we ask that lab heads send the Steering Committee a brief proposal describing their scientific goals related to the use of the dataset. This will give the existing community a clear idea of the specific cell populations or circuits that you are interested in and a sense for how much proofreading and annotation effort you can commit to the shared goal of producing a complete connectome. The goal is to have initial interests on paper to minimize project overlap, identify opportunities for collaboration, and make sure everyone is on the same page.
Proofreading is a skill, and true mastery of reconstructing neurons takes time and effort. We do not require mastery to pass the proofreading test but ask that all proofreaders focus on the “low hanging fruits” and only make edits that are clearly correct. This approach has been previously taken in the FAFB FlyWire dataset, and allowed for the collective expertise to increase as the "low hanging" edits were made. For more details about our proofreading workflow, see the README.
There are two stages in any proofreading project. The first is exploratory proofreading, which is used to identify neurons of interest in the dataset. The second is a formal proofreading project, which is more thorough and necessary for data ownership. It is important to clarify that data ownership does not imply absolute control over the data. However, it does protect your interests.
For initial proofreading efforts, we still expect individuals to follow guidelines for formal proofreading projects, though we anticipate less potential for overlap, as this is an important practice for maintaining a fair and collaborative community. Individuals should not be participating in more than one formal proofreading project at a time, but multiple people can be involved in the same project.
To find your neurons of interest, you will likely need to spend some time proofreading neuronal backbones. Some of these edits may enable formal proofreading projects, but many will not. These are still useful, as they will contribute to the collaborative community and may enable future projects by other groups. This is an important aspect of working in a collaborative community. You should not expect that exploratory proofreading will lead to authorship on papers related to those neurons unless you undertake a formal proofreading project related to them.
We are using the #proofreading-projects Slack channel to keep a record of what projects are ongoing and what has been completed. Please do not edit messages on Slack. Instead, post a reply to edit or correct a previous post.
In the initial phase of proofreading, projects will be assigned based on your proofreading team and scientific interests. If you are interested in a scientific question outside of the existing proofreading team projects, please contact the Steering Committee to discuss.
Within each proofreading team, there may be multiple ongoing proofreading projects. Team leads will make a post in the #proofreading-projects channel for each ongoing project that includes:
- a description of the project,
- the type of data that is being collected, and
- a list of the individual(s) involved in that project. At this point, the team lead and people working on that proofreading project have “preliminary ownership” of the data that will be collected.
Once the proofreading project has been completed, the team lead will make a comment on the initial post with a summary of the data that was collected, making sure to check the box next to “also send to #proofreading-projects”. This will finalize “full ownership” of the data. If a project has been abandoned, this will also be shared in a comment on the original post (and also sent to #proofreading-projects), terminating “preliminary ownership”.
Please note that anyone is allowed to proofread any neuron in the dataset, regardless of whether that neuron is part of a formal proofreading project or not. It is sometimes difficult to know if a neuron is part of another group’s project. If you realize that you have made an edit to a neuron that is part of a different proofreading project, we request that you send the relevant person(s) a Slack message.
There is a general moratorium/embargo on publishing (including preprinting) any studies that involve direct use of the dataset before the initial dataset preprint (planned to occur within 24 months of the start of grant, so by May 2027).
Individuals must have permission from the Steering Committee and the data owner(s) in order to share unpublished FAAB data at a conference, symposium, or with anyone outside of the community.
Users wishing to post preprints or submit manuscripts that involve direct use of the data before the embargo ends should seek prior approval from the Steering Committee. All those who helped to generate and develop the resources used should be offered co-authorship: a. By default, any work in the dataset is assumed to benefit from ongoing proofreading, annotation, and centralized coordination. Therefore, the producers of the aligned, autosegmented dataset and the co-PIs on the Wellcome grant would be co-authors. In addition, the contributions of all individuals who proofread and/or annotated any neurons used in the study should be evaluated and those individuals offered co-authorship or minimally acknowledgment. b. In the (less likely) case that the work does not make use of any previously proofread or annotated neurons, the producers of the aligned, autosegmented dataset and the co-PIs on the Wellcome grant would be co-authors. c. For edge cases of work not involving the autosegmentation or annotation, the producers of the aligned dataset would be co-authors. Once the initial dataset preprint has appeared, it should be cited in all future manuscripts that make use of the dataset. At that point, co-authorship requirements would revert to scientific norms, but studies relying on access to unpublished data prior to the preprint should still recognize the contributors to the resource with co-authorship. In addition, meaningful contributions to manuscripts by other dataset users should be recognized with co-authorship and otherwise handled in a manner consistent with scientific standards.
If a member's actions violate this Code of Conduct, access and privileges may be limited or revoked. Examples of violations include: unauthorized sharing with anyone who is not a member of the community, starting new projects or overtaking the projects of others without first discussing with the community, or repeated errors that cause ongoing problems for the community.
We expect to make modifications to this document. We also invite the community’s input concerning additions and/or modifications to these policies. Please contact the Steering Committee to offer suggestions. Ultimately, the Steering Committee reserves the right to make final decisions about proofreading and data ownership guidelines, but our main goal is to enable scientific progress by keeping the community harmonious, productive, and free of conflict.
This document was derived and adapted from the CRANTb community guidelines, the FANC community code of conduct, and the FAFB and FlyWire community guidelines.