Single cell analysis with "meta".tsv.gz file

Question

[Estevão]

I'm new to single-cell analysis and I would like to know how to produce the same UMAP plot from an article, since they provide both the h5 files and a .tsv.gz file with the following data:

UMAP_1  UMAP_2  Sample  Tissue  Treatment   Cluster nUmi    nGene   nUmiLog2    nGeneLog2
tumor_aPD1_AAACCCAAGCCTATTG 6.01140444140187    8.03710471733015    tumor_aPD1  tumor   aPD1    10  16215   3625    13.9850414033432    11.8237652797897

[Mensur Dlakic]

You are not providing enough information to help you properly - what paper and what figure?

Still, it seems that second and third columns contain UMAP XY coordinates, so you can use them to make a plot. It is up to you which of the remaining 8 columns to use for coloring the plot.

[Estevão]

Thank you very much, I will add more information below: The article and the GEO accession are these: Molgora M, et al. TREM2 Modulation Remodels the Tumor Myeloid Landscape Enhancing Anti-PD-1 Immunotherapy. Cell. 2020. GEO accession: GSE151710

The datasets I want to use are only: GSM4588939 and GSM4588940. I want to assemble the same UMAP as in supplementary figure S2, using the UMAP coordinates from this metadata file that I mentioned above. How should I do?

[ATpoint]

How should I do?

What is the question? The UMAP coordinates are present in that file, so I am unsure where you get stuck. Something like plot(x$UMAP1, x$UMAP2) is the simplest version.

[Estevão]

Thanks, it worked. To recreate the plot with the transcriptome data, can I do the Seurat pipeline and then plot with the coordinates?