Error in plot.window(...) : need finite 'xlim' values

[2021JohnSheng]

Hi,
I am encountering a similar issue to the https://github.com/mrb20045/CaSilico/issues/3
Error in plot.window(...) : need finite 'xlim' values
Interestingly, this issue does not occur on a Windows system; the code executes without any errors there.
To address this issue on Linux, I added the following lines to my script to ensure the software continues running even if this error occurs:
run_CaSilico_with_error <- function(file_path) {
tryCatch({
run_CaSilico(file_path)
}, error = function(e) {
if (grepl("Error in plot.window", e$message)) {
message("Plotting error ignored: ", e$message)
} else {
stop(e)
}
})
}

However, despite this attempt, the software fails to produce the final output files, such as Spacers_Information.Html and Off_Targets_Information.Html.

Could you please provide guidance on how to resolve this issue? Any assistance would be greatly appreciated.

Thank you.

[xingqi-Niu]

Hi,

I'm encountering exactly the same issue described in #3, but with different circumstances and more severe consequences. The Error in plot.window(...) : need finite 'xlim' values occurs on my Windows 10 system (contrary to #3 where it only happened on Linux). The critical issue is that after this error, no HTML output files are generated at all.

I've tried several approaches: using tryCatch() to ignore plotting errors (similar to #3), lowering ConservationThreshold values (from 0.98 to 0.90), trying different CRISPR types (casVI_A, casVI_B), and using the example data (3D.fasta). The software continues running after the error, but it never produces Spacers_Information.Html or Off_Targets_Information.Html.

Currently, the software creates folders (like casVI_A-ConservationMethod_1-threshold_98-result_Test) containing subfolders such as Multiple Sequenc Alignment, Plots, and Secondary Structure. Sequence alignment seems to work (there are .aln files), but all computed results (crRNA designs, scores, off-target analysis) are lost because HTML generation never happens.

My actual need is to score pre-designed crRNAs using local alignment files without visualization. I'm wondering if there's: 1) a way to skip plotting entirely (e.g., a plot = FALSE parameter), 2) a minimal working version that focuses only on computation, or 3) any workaround to extract computed data before the crash.

System Information:
OS: Windows 10 Professional
R version: 4.3.2
RStudio: 2023.12.0
CaSilico: Latest from CRAN (1.0.0)
Dependencies: seqinr, DT, RRNA, ape, htmltools, etc. all installed
Reproduction steps:
library(CaSilico)
data <- paste0(system.file(package = "CaSilico"), "/data/3D.fasta")
CaSilico(ResultsFolder = "Test",
TargetFasta = data,
CrisprTypes = "casVI_A",
OffTarget = FALSE)

Thank you for your work on this package. Any guidance would be greatly appreciated.