Help with fugossem

Hello! I am trying to use fugossem to characterize some hypothetical proteins. Can the input MTX file include samples as columns and gene names as rows from a transcriptomic analysis or does it have to be from a proteomics experiment? Then does the GO tsv have to be the associated GO for some of the known genes (does not need all)? Something like below (imagine it’s a matrix though)

GeneSymbol Sample 1 Sample 2 Sample 3 Sample 4 Sample 5 Sample 6 Sample 7 Sample 8 Sample 9
AW25_RS00005 657.015162978176 727.518425755497 762.100473024637 647.636291632532 709.104354077248 647.471934451004 402.94912816652 401.721369224263 498.328664648842
AW25_RS00010 2580.74234603125

Hi there,

Thanks for your interest in FUGAsseM. Regarding your questions:

  1. MTX input file: Expression data from a transcriptomics experiment should work. The matrix can contain samples as columns and genes as rows.
  2. GO annotation file: This should include GO annotations for the known genes. It does not need to cover all genes—only those with available annotations.

We provide demo examples of all the required input files on the tutorial page. Please take a look there for more detailed guidance on the expected formats: FUGAsseM · biobakery/biobakery Wiki · GitHub

Best,
Yancong

Hi Yancong,

Thank you for your help. Once I run the analysis, what files are the most helpful to discern hypothetical proteins and functions? There are a lot of output files and I’m not sure where to begin.

Thank you,

Hiral

Hi Hiral,

The prediction results from all taxa will be organized in the $OUTPUT_NAME/merged directory, with filenames formatted like $BASENAME.finalized_ML.prediction.tsv. More details about the output files are available on our tutorial page:

Our tutorial page provides detailed explanations of the input and output files. I would recommend starting with the demo run described there—this can help you become familiar with the workflow and better understand how to interpret the results generated by the tool.

Thanks!
Yancong