Unusual graph for Identifying HVGs from the normalized log-expression
Entering edit mode
es.azadian • 0
Last seen 3.6 years ago

Hello to everyone,

I've followed the "A step-by-step workflow for low-level analysis of single-cell RNA-seq data with Bioconductor, V2" for my own data. The problem is that when I produce the graph for Variance of normalized log-expression values (using trendVar and decomposeVar functions), my plot (an ascending-linear shape in which all of dots are accumulated and correlated with the line of y=x, and it's not at all a bell-shaped graph) is basically different with the plot of workflow (Figure 10). Is it normal?!

I'm wondering if you could give me some hints.



workflows simplesinglecell • 461 views
Entering edit mode
Aaron Lun ★ 27k
Last seen 3 hours ago
The city by the bay

I'm not a telepath so these issues are very difficult to diagnose without code and plots (see the posting guide). Nonetheless, there are two obvious possibilities that come to mind.

  • You're computing the mean-variance trend on the raw counts instead of the log-transformed normalized counts. For Poisson-distributed counts, this should give a straight line.
  • You have run the code correctly but your counts are so low that you are effectively only visualizing the linear part of the plot. This would not be impossible for very low-coverage droplet-based data.

In any case, I would suggest following the updated instructions at http://bioconductor.org/packages/devel/workflows/html/simpleSingleCell.html rather than reading the article, as almost all of the code in the latter is out of date.


Login before adding your answer.

Traffic: 199 users visited in the last hour
Help About
Access RSS

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6