I have gone through the tutorial at MERFISH whole brain spatial transcriptomics (part 1) — Allen Brain Cell Atlas - Data Access. I understand that the input data should include x, y, and z coordinates. Could you clarify what other information is required? For example, is the cluster_alias information essential?
In my case, I have not performed any cell clustering based on RNA expression. However, I would like to annotate the different brain regions for my whole brain dataset. Is this possible?
I would appreciate your guidance.
Thank you in advance.
@jeremyinseattle @tylermo Can you please help me with this?
Hi @kulansam
There are three different topics here that I want to separate.
- Data access and exploration of Allen Institute MERFISH data. If this is what you want to do, then the tutorial above is the correct starting point. I don’t think that is the best approach for your specific use case.
- Assigning Allen Institute cell types to user-provided MERFISH data. This is the exact use case of MapMyCells. If this is what you want to do, I’d suggest saving your MERFISH data to an h5ad file and mapping to the whole mouse brain taxonomy by following these steps.
- Assigning brain regions to user-provided MERFISH data. If this is what you want to do, I’d suggest starting with this newer tutorial on how we have aligned Allen Institute MERFISH data to the Allen common coordinate framework (CCF). This forum post includes several other related resources.
If you can’t find what you need in the response above, please provide additional details about your use case.
Best,
Jeremy
Hi @jeremyinseattle,
Thank you so much for your clear response. I really appreciate it!
My work is focused on the third approach, where I aim to map my MERFISH slice onto the Allen Brain Atlas and annotate the regions. I went through the tutorial (MERFISH CCF mapped coordinates — Allen Brain Cell Atlas - Data Access), but I wasn’t clear on the format for the input file.
For my own MERFISH data, the metadata file contains the following columns:
EntityID, fov, volume, center_x, center_y, min_x, min_y, max_x, max_y, anisotropy, transcript_count, perimeter_area_ratio, solidity, DAPI_raw, DAPI_high_pass, PolyT_raw, PolyT_high_pass.
However, I don’t have Z-coordinates/z_section, average_correlation_score, cluster_alias etc. information (as mentioned in the tutorial’s example file).
It would be great if you could help me understand how to prepare the input file.
Thanks again for your help!
Unfortunately, we don’t have any MERFISH data on developing mouse brain data, and the type of registration you are requesting is not something scientists at the Allen Institute have done yet, so the folks that I asked do not have answers. I’d encourage others in the community to reply if you can help!
One other potential place is look is at the Illustrated developing mouse brain anatomic atlases found here. It’s possible that some manual annotation could be done using this(?).
Thank you for your response. Based on your previous message, you mentioned three different options, and option 3 is the one I would like to pursue. I feel OK with registering my MERFISH slice by comparing it to the adult mouse (P56) atlas. I believe there will not be much difference between P6 and P56. Hence, could you please provide me with the necessary information on preparing the input files?
For manual annotation, the Allen Atlas sections for P4 and P14 are sagittal, while my sections are coronal. Therefore, I believe it will be challenging to map them directly.
Is there any advice on how to proceed (brain region annotations) would be greatly appreciated.
As of right now, we use some code internal to the Allen Institute for aligning adult MERFISH data to the common coordinate framework and assigning brain regions. A method for training a classifier to assign brain regions directly to cells is being developed, but isn’t ready at this time.
In addition to the post above, this post and this list of community resources may be of some value.