Hi all, are there sample CSVs available for what an input CSV directly going into the MICOM build step might look like? After inputting species level data (Based on kraken2 Refseq 2023 DB assignments) into the build step, I find that there are zero database matches to the AGORA103 database.
This is based on the examples provided at https://github.com/Gibbons-Lab/scfa_predictions.
Attaching a screenshot of how I formatted my data before going into the tutorial code above. If there is an example of exactly what format MICOM (Refseq Database format) expects the data to be in when going into the build step, that would be super helpful. It is difficult to view the .qza database and know which parts MICOM is using to match taxonomy.
Separately, I've succeeded in building models with the Genus level (GTDBv207) data and it's corresponding UHGG-GTDBv207 database, but with the disadvantage that the scfa_prediction paper's downstream grow() steps could not be run, since the substrate name formats are different.
Thanks so much!
Hi all, are there sample CSVs available for what an input CSV directly going into the MICOM build step might look like? After inputting species level data (Based on kraken2 Refseq 2023 DB assignments) into the build step, I find that there are zero database matches to the AGORA103 database.
This is based on the examples provided at https://github.com/Gibbons-Lab/scfa_predictions.
Attaching a screenshot of how I formatted my data before going into the tutorial code above. If there is an example of exactly what format MICOM (Refseq Database format) expects the data to be in when going into the build step, that would be super helpful. It is difficult to view the .qza database and know which parts MICOM is using to match taxonomy.
Separately, I've succeeded in building models with the Genus level (GTDBv207) data and it's corresponding UHGG-GTDBv207 database, but with the disadvantage that the scfa_prediction paper's downstream grow() steps could not be run, since the substrate name formats are different.
Thanks so much!