Hi! First, I just wanted to thank you for sharing all of this data with the community! My current team and I are working on a new algorithm and using your dataset as one of your trial runs. We currently are using the filtered and processed metadata file for the cells IDs identified in the TAG column (syn18686381) with the gene row names (syn18686382) to construct the fully annotated count matrix. In your metadata file (syn18686381), you also include both broad cell type and subcluster information. However, I was wondering if you also had the information to match each cell ID (TAG) to its respect pathology from your analysis (no pathology, early AD pathology, and late AD pathology)? Any direction and/or help would be much appreciated! Thanks, Amna

Created by Amna Irfan airfan1
Hi! I also have the same issue than Amna. I see how we can map the specimenIDs with the projids and their clinical metadata. However, how can we assign each cell to their specimenIDs? Thanks! Violeta
Hi Amna, The biospecimen metadata file for this study (syn18642936) should let you map specimenIDs from the data to ROSMAP donor projids, and you can use the projids to map to individual phenotype data in the ROSMAP clinical metadata file at syn3191087. The ROSMAP clinical file does not actually have categories for AD diagnosis, but it does have Braak stages, CERAD scores, cognitive diagnoses, and other variables available for each donor (codebook is here: syn3191090). The criteria based on these variables used for diagnosis in a particular analysis is hopefully provided by the authors of the analysis! Best, Abby

Matching cell IDs with AD pathology page is loading…