This repository keeps track of the datasets found in the literature and the efforts to obtain them. So far, I found the following datasets that could be of interests for the ASAP microbiome analytics core. The strategy to find them so far was to review the results of the following Google scholar search. In the future, a double check by other researchers on different paper browsers (e.g. Scopus, WebOfScience) would strengthen the systematicity.
With the help of Svetlana (@SvetlanaUP) and the Outreachy contributors, we are adding 8 new shotgun datasets to BugSigDB.
| Item | Study | GitHub issue | BugSigDB Study | Status |
|---|---|---|---|---|
| 1 | LiZ_2021 | waldronlab/BugSigDBcuration#600 | https://bugsigdb.org/34276644 | curated and reviewed |
| 2 | NieS_2023 | waldronlab/BugSigDBcuration#601 | https://bugsigdb.org/36995383 | curated and reviewed |
| 3 | PalaciosN_2023 | waldronlab/BugSigDBcuration#602 | https://bugsigdb.org/37314861 | curated and reviewed |
| 4 | DuruIC_2024 | waldronlab/BugSigDBcuration#603 | https://bugsigdb.org/39143178 | curated and reviewed |
| 5 | Metcalfe-RoachA_2024 | waldronlab/BugSigDBcuration#604 | https://bugsigdb.org/39192744 | curated and reviewed |
| 6 | NishiwakiH_2024 | waldronlab/BugSigDBcuration#605 | https://bugsigdb.org/38773112 | curated and reviewed |
| 7 | NuzumND_2023 | waldronlab/BugSigDBcuration#606 | https://bugsigdb.org/37150399 | curated and reviewed |
| 8 | BolliriC_2022 | waldronlab/BugSigDBcuration#607 | https://bugsigdb.org/35852145 | curated and reviewed |
Summary of PD studies selection and curation process in BugSigDB:
Papers selected by @g-antonello: 16
Preprints (BugSigDB does not accept preprints): 3
Papers with no significant differential abundance results to curate: 2
BugSigDB already curated: 3
BugSigDB additionally curated by Kate Rasheed, Svetlana Ugarcina Perovic and bugsigdb reviewers: 8
Our protocol for a flowchart of the selection process for eligible articles in the PubMed database:
Identification: https://pubmed.ncbi.nlm.nih.gov/?term=parkinson%2C+microbiome%2C+microbiota
Screening: manual screening for papers not already included in BugSigDB (until July 2024) with significant differential abundance results
Included: 42 papers <- sharing a list of our PD curation efforts by Outreachy contributors and Svetlana (since July 2024): https://github.com/waldronlab/BugSigDBcuration/issues?q=is%3Aissue%20state%3Aclosed%20parkinson%20label%3Areviewed
NOTE: we selected both amplicon and whole genome sequencing studies.
The work year has ended with 2 contacts active to obtain metadata by
NishiwakiH_2024 and MetcalfeRoachA_2024.
So far, we have downloadable Fastq files and retrievable case-control data for the following datasets:
- BedarfJR_2017 (Alice Bassetto shared it)
- MaoL_2021 (Alice Bassetto shared it)
- JoS_2022 (case-control in biosample titles in SRA)
- WallenZD_2022 (Haydeh’s public dataset)
- BoktorJC_2023 (2 datasets, metadata in Supplementaries)
- ZhangM_2023 (all PD, different sex)
- LeeEJ_2024 (all PD, different sub-types)
- NishiwakiH_2024 (the only ones to share metadata after emailing them)
- MetcalfeRoachA_2024 (emailing in progress)
Datasets more likely to be obtained may also be:
- BeckerA_2022 (available in Chinese database, I am requesting access)
- NieS_2023 (available in Chinese database, I am requesting access)
- PalaciosN_2023 (data in dbGAP, Curtis HUttenhower is last author), but this exploits data from Nurses study in Boston, data sharing is harder.
- StagmanK_2024 (deposited in MJFF, I requested full access)
Since we are having trouble with responses by authors, we decided to expand
our search towards other -omic data, which would still make a valuable addition
to the field of PD. Our attention will focus first on 16S amplicon sequencing.
For this search, we created a search on BugSigDB for studies on PD in H. sapiens
with Condition: parkinson's disease | parkinsonism | parkinson | parkinson disease.
Whenever a shotgun study is found that is not yet in parkinson_shotgun_datasets.tsv,
it is added to that file. 16S studies will be annotated with the same logic
in a different file, named parkinson_16S_datasets.tsv. My week's efforts will
mostly be around curating these datasets.
A new email version has been developed:
Dear Dr. **Last Name**,
I am working on a project funded by the "Aligning Science Across Parkinson" (ASAP) network and the Michael J. Fox foundation to build a harmonized resource of all publicly available metagenomic data available for the study of the gut microbiome in Parkinson's disease. We are following the approach of the curatedMetagenomicData (cMD) initiative (Pasolli et al., 2017) to lower barriers to reuse and integrated analysis by the entire Parkinson's Disease research community to better understand the underlying causes of the disease.
To succeed, we need access to raw sequencing data and individual participant metadata, including the datasets you have generated **https://doi.org/link**. We will uniformly process raw data through the latest version of bioBakery (https://huttenhower.sph.harvard.edu/biobakery_workflows/), and standardize participant data across all studies. The minimum data we need in order to include your study is the raw reads and Parkinson's disease/health status, but age and sex are important as well. Ideally, it would be best to share the full data set you worked with, for the sake of completeness, reusability, and reproducibility of results.
I am confident you will understand the importance of this harmonization effort and will be part of it as a data contributor. I'm looking forward to discussing details, including addressing any limitations or concerns you have about data sharing.
Giacomo Antonello, PhD
Microbiome Analytics Core (MAC), ASAP Network
MAC GitHub: https://github.com/ASAP-MAC
Book a meeting with someone at MAC: https://outlook.office365.com/book/[email protected]/
Finished First shotgun datasets list, containing 15 datasets.
So far, the request of metadata has been piloted on only 3 research groups, with no success. I sent the emails from my private email account, Cc-ing Levi, to the corresponding author, and when not the same, to the last author too. This was the email template, which I adapted to each specific case:
Dear <corresponding author>,
I am writing on behalf of the Curated Metagenomic Data (cMD) project (Pasolli et al., 2017) led by prof Levi Waldron.
We are currently working on a new data release focused on Parkinson’s disease in a project funded by the "Aligning Science Across Parkinson" (ASAP) network and the Michael J. Fox foundation.
We appreciated reading your contribution in <Journal Name> this year (<doi link>) and for making the sequencing data publicly available. We would like to include your data set in our next cMD release. To do so, we would need have access to individual level metadata. The bare minimum is disease/health status, but age and sex are important as well.
Ideally, it would be best if you could provide us with the full data set you worked with, both for the sake of completeness, reusability, and reproducibility of data, not to mention the higher visibility your contribution would gain.
I am looking forward to discussing the details together.
Best regards
Giacomo Antonello, PhD
Metadata retrieval has been attempted using a handy manual search tool called SRA-explorer, which provides a nice interface to download metadata and create wget downloading commands for FASTQ files.