search

ctlab / fgsea

Fast Gene Set Enrichment Analysis
Other
379 stars 67 forks source link

is there a plan to implement fast ssGSEA? #117

Closed Yunuuuu closed 2 years ago

Yunuuuu commented 2 years ago

We often use ssGSEA to calculate scores based on specific gene set, when we want to compare these different scores corresponding to different gene sets related to specific function, it's necessary to normalize these scores then compare them since the size of the underlying gene sets of these scores can give these scores different weight. Though the broadinstitute provide these function in https://github.com/broadinstitute/ssGSEA2.0, it's very slow and spend a lot of times to run a work.

Thanks to fgsea, we can run a lot of works easily and quickly, it would be great if fgsea can also run ssGSEA.

assaron commented 2 years ago

Did you try GSVA package? It has a relatively fast ssGSEA function implementation, but not sure how it compares with Broad's.

Yunuuuu commented 2 years ago

Thanks for your reply @assaron, GSVA do provide a fast ssGSEA but cannot caculate the NES (normalized enrichment scores), just like the regular GSEA. so we cannot compare different scores in a sample (For example: score based on gene set Macrophage vs score based on Neutrophil). We can only compare the specific scores related to a specific gene set (like score based on Macrophage gene set) between different samples.

assaron commented 2 years ago

Still I think it's probably better to ask @rcastelo for this feature to be added to GSVA. Our lab doesn't normally use ssGSEA, so we're unlikely to add such functionality to fgsea package.