dviraran
commented
4 years ago Hi,
Yes, you are completely correct about the mappings in Blueprint. While it is quite this a subject of undergoing debate, alternatively activated macrophages is an in vitro method to characterize M2 (see https://en.wikipedia.org/wiki/Macrophage_polarization for example).
I am attaching the mappings for Blueprint and HPCA that were used for the development of xCell.
blueprint_samples.txt hpca_samples.txt
Hope this helps.
Best, Dvir
ikalatskaya
Our group extensively uses xCell for some research projects. Thank you very much for sharing this package with community, it works smoothly and generates reliable results. The only issue that we have is with result interpretation in some cases. For example, for macrophage signatures (M1/M2). Taking into account that macrophages are generally tissue resident and not part of the blood stream, it is essential for us to know how the samples used for signature generation were produced. That’s why we would truly appreciate if you can clarify the origin of M2 signatures in your xCell package.
You are claiming that M2 signature consists of six subsignatures: three of those were from BLUEPRINT and three from HPCA (Mabbott et al, 2013). In BLUEPRINT macrophage samples are not labeled M2 vs M1. It would be great if you can confirm that “alternatively activated macrophage” from BLUEPRINT were selected for M2 signature development. It would be very helpful if you can provide exact ERX sample ids that were used. Some of them were isolated from venous blood, and some – from cord blood, which makes the interpretation of the deconvolution results very challenging. In addition, there are over 90 samples marked as “macrophages” in HPCA study. How did you select M2?
Thank you in advance!