Hi everyone, I'm new to single cell so I apologize if any of this is due to my experience. I had previously used a prior version of harmony without much trouble (unfortunately I don't remember which version it was), but since updating to 1.1.0 I've had some issues. When trying to integrate two small Seurat single-cell datasets (both are <10000 cells), I run the following command:
But this stops at 2/10 harmony iterations. Looking at the convergence plot I inferred that they hadn't converged properly.
So I run the command for harmony_options(epsilon.cluster = -Inf, epsilon.harmony = -Inf) which I believe runs successfully, and change the last line of the above code as follows:
Theoretically, this would prevent the harmony process from ending early until it reaches 20 iterations, but even after adding those parameters RunHarmony still converges at 2/10 iterations. Changing the epsilon.cluster/epsilon.harmony values to anything else also yields to the same result.
But this process returns the following and stops after 2/10 harmony iterations:
Transposing data matrix
Initializing state using k-means centroids initialization
Harmony 1/20
0% 10 20 30 40 50 60 70 80 90 100%
[----|----|----|----|----|----|----|----|----|----|
**************************************************|
Harmony 2/20
0% 10 20 30 40 50 60 70 80 90 100%
[----|----|----|----|----|----|----|----|----|----|
**************************************************|
Harmony converged after 2 iterations
Warning: Invalid name supplied, making object name syntactically valid. New object name is Seurat..ProjectDim.RNA.harmony; see ?make.names for more details on syntax validity
Is there a workaround for this problem? I know I saw another issue with ignored parameters but I believe it may already be covered with this 1.1.0 update.
Hi everyone, I'm new to single cell so I apologize if any of this is due to my experience. I had previously used a prior version of harmony without much trouble (unfortunately I don't remember which version it was), but since updating to 1.1.0 I've had some issues. When trying to integrate two small Seurat single-cell datasets (both are <10000 cells), I run the following command:
But this stops at 2/10 harmony iterations. Looking at the convergence plot I inferred that they hadn't converged properly. So I run the command for harmony_options(epsilon.cluster = -Inf, epsilon.harmony = -Inf) which I believe runs successfully, and change the last line of the above code as follows:
Theoretically, this would prevent the harmony process from ending early until it reaches 20 iterations, but even after adding those parameters RunHarmony still converges at 2/10 iterations. Changing the epsilon.cluster/epsilon.harmony values to anything else also yields to the same result. But this process returns the following and stops after 2/10 harmony iterations:
Is there a workaround for this problem? I know I saw another issue with ignored parameters but I believe it may already be covered with this 1.1.0 update.
sessionInfo():