Dear all, I downloaded Biocrates P180 kit metabolomic data from AMP-AD (ROSMAP) and M2ove-AD (ADMC) projects. They have indicated how many samples have been analysed in each of these projects. However, I was unable to find further information regarding the samples: whether all the analyzed samples belong to the disease group or there are samples from different classes (Controls, Cognitive impairment and AD)? Can you guide me on where I can find the information regarding sample classification? Thanking you, Best regards, Sai.

Created by Sai Prashanth Santhapuri Santhapuri
Excellent, I'm glad to hear it! Best of luck with your research.
Hello Abby, Sorry for the delay. I want to bring to your kind notice that finally using the information from both the ROSMAP Clinical Codebook (pdf file) & ROSMAP Clinical (.csv file) I was able to merge the individual IDs and their respective diagnosis. It is now easy to identify the diagnoses of the individual IDs of the ROSMAP study. Thank you very much for your valuable assistance. Thanking you, Yours sincerely, Sai.
For ROSMAP, the p180 data file has a column called "projid", which is a Rush identifier (I believe an abbreviation for "project ID") for each donor. You can map this "projid" column in the data file to the "projid" column in the clinical metadata file, which should let you link the data to diagnosis info.
Hello Abby, I hope you are keeping well. Thank you very much for the resourceful information. It worked out for the ADMC_UPenn data. As you indicated, they have provided the **individual IDs and respective diagnoses in the phenotype metadata file**. I checked both the **"Biospecimen and Clinical metadata"**, however, I am still unable to find something similar w.r.t ROSMAP data, unlike ADMC_UPenn. Your help will be highly appreciated. Thanks a lot once again, Best regards, Sai.
Oh I see, sorry for the misunderstanding. In order to match the sample IDs in the data (.csv) files with metabolite quantities, such as the ROSMAP p180 data file you mentioned, you will need to join the metadata file to the data file. For the ROSMAP data file, you can join it to the biospecimen metadata (specimenID = Sample.Identification) or to the clinical metadata (projid = projid). For the ADMC_UPenn data, you can join the p180 data file to the phenotype metadata on the "INDDID" column. Unfortunately these two studies were contributed to the Portal before we had standardized our metadata, so the joining columns are a little unclear. Does this help? Happy to help troubleshoot further!
Dear Abby, Thanks for your valuable response. As indicated, yes there is a column which says " 'Dx' refers to diagnosis, either AD/MCI/NCI" however in a working file (.csv), for example here is an excel sheet I am working on **"4688UPLCP180DATA.csv"**, if you look at the data provided in the mentioned excel sheet, it does not give any information on diagnosis. Metadata files have given descriptions of the abbreviations used in the excel sheets. However, I am looking for an excel sheet which can help in matching the IDs in the **.csv files** and their respective diagnosis. Sorry, I am actually confused in finding such a file, thus too many queries. I thank you for patiently answering my queries, I hope I will be able to find it soon. Thanking you, Best regards, Sai.
Hi Sai, According to the [data dictionary](https://www.synapse.org/#!Synapse:syn17316129) for the UPenn ADMC phenotype data, there should be a column called "Diagnosis" with the consensus or clinical diagnosis of AD/MCI/NCI. For the ROSMAP data, there are values representing Braak and CERAD scores for individuals, as well as a variable called 'cogdx' which has the final consensus cognitive diagnosis. The [ROSMAP clinical codebook](https://www.synapse.org/#!Synapse:syn3191090) has a key that maps the numeric values to AD/MCI/NCI diagnoses. Does that help you find the information you need? Best, Abby
Dear Abby, Thank you very much for your response. I downloaded the metadata files, however, they have given information about various aspects including : 1: Sex 2: Education 3:Race So on., but they have not mentioned if the individual has been diagnosed with AD or he is a control subject, etc. Can you direct me to find such information? Your response is highly appreciated. Thanking you, Best regards, Sai.
Hi Sai, [This query](https://adknowledgeportal.synapse.org/Explore/Data?QueryWrapper0=%7B%22sql%22%3A%22SELECT%20*%20FROM%20syn11346063.25%22%2C%22limit%22%3A25%2C%22offset%22%3A0%2C%22selectedFacets%22%3A%5B%7B%22concreteType%22%3A%22org.sagebionetworks.repo.model.table.FacetColumnValuesRequest%22%2C%22columnName%22%3A%22study%22%2C%22facetValues%22%3A%5B%22ROSMAP%22%2C%22ADMC_UPenn%22%5D%7D%2C%7B%22concreteType%22%3A%22org.sagebionetworks.repo.model.table.FacetColumnValuesRequest%22%2C%22columnName%22%3A%22resourceType%22%2C%22facetValues%22%3A%5B%22metadata%22%5D%7D%2C%7B%22concreteType%22%3A%22org.sagebionetworks.repo.model.table.FacetColumnValuesRequest%22%2C%22columnName%22%3A%22dataType%22%2C%22facetValues%22%3A%5B%22metabolomics%22%2C%22clinical%22%5D%7D%5D%7D) shows all the metadata files from the ROSMAP and ADMC_UPenn studies in the AD Knowledge Portal that have clinical or metabolomics metadata. For the ADMC_Upenn study, there is a phenotype file and a data dictionary for that phenotype, as well as some protocols and data dictionaries for the assay. For ROSMAP, you can join the specimenIDs in the Biocrates data to the biospecimen metadata file, which can then be joined to the ROSMAP clinical file to get all of the available clinical metadata, including diagnosis. Further documentation about how to join AD Portal metadata files is available [here](https://help.adknowledgeportal.org/apd/About-Metadata.2241626149.html#AboutMetadata-Metadatarequirements). Let me know if that helps -- happy to answer any other questions! Abby

Information regarding sample classification page is loading…