error in mimseq: DESeq cannot estimate dispersions

[kelly-sovacool]

/scratch/cluster_scratch/sovacoolkl/tranquil_11-13/logs/44453939.44454181.mimseq.contrast=Tet1n2DKOalkBtRNA_vs_WTtRNA.err

estimating size factors
estimating dispersions
gene-wise dispersion estimates
mean-dispersion relationship
Error in estimateDispersionsFit(object, fitType = fitType, quiet = quiet) :
  all gene-wise dispersion estimates are within 2 orders of magnitude
  from the minimum value, and so the standard curve fitting techniques will not work.
  One can instead use the gene-wise estimates as final estimates:
  dds <- estimateDispersionsGeneEst(dds)
  dispersions(dds) <- mcols(dds)$dispGeneEst
  ...then continue with testing using nbinomWaldTest or nbinomLRT
Calls: DESeq ... estimateDispersions -> .local -> estimateDispersionsFit
In addition: Warning messages:
1: In DESeqDataSet(se, design = design, ignoreRank) :
  some variables in design formula are characters, converting to factors
2: In DESeqDataSet(se, design = design, ignoreRank) :
  some variables in design formula are characters, converting to factors
Execution halted

[kelly-sovacool]

improved error message with exact command

+----------------------------------------------+
| Differential expression analysis with DESeq2 |
+----------------------------------------------+
2024-11-19 08:53:33,463 [INFO ] ['Rscript', '/scratch/cluster_scratch/sovacoolkl/mim-tRNAseq/mimseq/deseq.R', '/scratch/cluster_scratch/sovacoolkl/tranquil_11-13/results/Tet1n2DKOalkBtRNA_vs_WTtRNA/mimseq_11-19/', '/scratch/cluster_scratch/sovacoolkl/tranquil_11-13/results/Tet1n2DKOalkBtRNA_vs_WTtRNA/mimseq_11-19/sampleinfo_cov.txt', 'WTtRNA', '0.95', '0.05', '/scratch/cluster_scratch/sovacoolkl/mim-tRNAseq/mimseq/data/hg38-eColitK/hg38-mitotRNAs.fa']
Warning messages:
1: package ‘DESeq2’ was built under R version 4.2.3
2: package ‘S4Vectors’ was built under R version 4.2.3
3: package ‘IRanges’ was built under R version 4.2.3
4: package ‘GenomicRanges’ was built under R version 4.2.3
5: package ‘GenomeInfoDb’ was built under R version 4.2.3
6: package ‘matrixStats’ was built under R version 4.2.3
7: package ‘Biobase’ was built under R version 4.2.3
Warning message:
package ‘RColorBrewer’ was built under R version 4.2.3
Warning message:
package ‘pheatmap’ was built under R version 4.2.3
Warning message:
package ‘ggplot2’ was built under R version 4.2.3
Warning messages:
1: package ‘calibrate’ was built under R version 4.2.3
2: package ‘MASS’ was built under R version 4.2.3
Warning message:
package ‘plyr’ was built under R version 4.2.3
Warning message:
package ‘dplyr’ was built under R version 4.2.3
Warning message:
package ‘circlize’ was built under R version 4.2.3
estimating size factors
estimating dispersions
gene-wise dispersion estimates
mean-dispersion relationship
Error in estimateDispersionsFit(object, fitType = fitType, quiet = quiet) :
  all gene-wise dispersion estimates are within 2 orders of magnitude
  from the minimum value, and so the standard curve fitting techniques will not work.
  One can instead use the gene-wise estimates as final estimates:
  dds <- estimateDispersionsGeneEst(dds)
  dispersions(dds) <- mcols(dds)$dispGeneEst
  ...then continue with testing using nbinomWaldTest or nbinomLRT
Calls: DESeq ... estimateDispersions -> .local -> estimateDispersionsFit
In addition: Warning messages:
1: In DESeqDataSet(se, design = design, ignoreRank) :
  some variables in design formula are characters, converting to factors
2: In DESeqDataSet(se, design = design, ignoreRank) :
  some variables in design formula are characters, converting to factors
Execution halted
2024-11-19 08:53:44,432 [INFO ] DESeq2 outputs located in: /scratch/cluster_scratch/sovacoolkl/tranquil_11-13/results/Tet1n2DKOalkBtRNA_vs_WTtRNA/mimseq_11-19/DESeq2

so the Rscript command is

Rscript /scratch/cluster_scratch/sovacoolkl/mim-tRNAseq/mimseq/deseq.R \
  /scratch/cluster_scratch/sovacoolkl/tranquil_11-13/results/Tet1n2DKOalkBtRNA_vs_WTtRNA/mimseq_11-19/ \
  /scratch/cluster_scratch/sovacoolkl/tranquil_11-13/results/Tet1n2DKOalkBtRNA_vs_WTtRNA/mimseq_11-19/sampleinfo_cov.txt \
  WTtRNA 0.95 0.05 \
  /scratch/cluster_scratch/sovacoolkl/mim-tRNAseq/mimseq/data/hg38-eColitK/hg38-mitotRNAs.fa

but sampleinfo_cov.txt is empty, but it exists in the cov/ subdir (results/Tet1n2DKOalkBtRNA_vs_WTtRNA/mimseq_11-19/cov/sampleinfo_cov.txt)

[kelly-sovacool]

error with improved entracing:

Error in estimateDispersionsFit(object, fitType = fitType, quiet = quiet) :
  all gene-wise dispersion estimates are within 2 orders of magnitude
  from the minimum value, and so the standard curve fitting techniques will not work.
  One can instead use the gene-wise estimates as final estimates:
  dds <- estimateDispersionsGeneEst(dds)
  dispersions(dds) <- mcols(dds)$dispGeneEst
  ...then continue with testing using nbinomWaldTest or nbinomLRT
Calls: DESeq ... estimateDispersions -> .local -> estimateDispersionsFit
In addition: Warning messages:
1: In DESeqDataSet(se, design = design, ignoreRank) :
  some variables in design formula are characters, converting to factors
2: In DESeqDataSet(se, design = design, ignoreRank) :
  some variables in design formula are characters, converting to factors
Backtrace:
    ▆
 1. └─DESeq2::DESeq(dds_anticodon)
 2.   ├─BiocGenerics::estimateDispersions(...)
 3.   └─DESeq2::estimateDispersions(...)
 4.     └─DESeq2 (local) .local(object, ...)
 5.       └─DESeq2::estimateDispersionsFit(object, fitType = fitType, quiet = quiet)
Execution halted

[kelly-sovacool]

This issue only occurs with our test dataset but does not occur with mim-tRNA-seq's test dataset, nor with Colin's real data. We should switch to using the test data bundled with mim-tRNA-seq.