Error in `geom_point()`

[handesome]

hi！Thank you for developing the wonderful package!!when I run the example code grid.arrange(grobs = trajectory_plots, layout_matrix=matrix(c(1:2), 1, 2)) Error in geom_point(): ! Problem while computing aesthetics. Error occurred in the 1st layer. And in the p1 <- plot_cycle_diff_3d(normalizedResultLst, redpath_pseudotime, cell_cycle_labels, cell_type_labels), how to get the normalizedResultLst? Thanks!

[kenxie7]

Hi there!

Thanks for your interest in the package!

For the trajectory plots, could you please check if the variable "trajectory_plots" is empty or does it give you the right plots? I am unsure whether it's the grid.arrange function that might be giving an error here.

The normalizedResultLst is part of the output after running reCAT (tinglabs/reCAT), and it gets saved with this code "save(finalHighCorLst, highCorLst, resultLst, normalizedResultLst, ensembleResultLst, file = paste("bestEnsembleComplexTSP", beginNum, "-", endNum, name, ".RData"))" by default.

After you run get_ordIndex from the reCAT package, you should see a .RData file beginning with "bestEnsembleComplexTSP", and you can load the objects from there.

Best, Ken

[handesome]

Hi! Thank you for the quick reply! For the trajectory plots, the variable "trajectory_plots" is shown and the codegrid.arrange(grobs = trajectory_plots, layout_matrix=matrix(c(1:2), 1, 2))give the error Error ingeom_point(): ! Problem while computing aesthetics. Error occurred in the 1st layer. Caused by error inunique.default(): ! unique()只适用于矢量 Runrlang::last_trace()to see where the error occurred. what should I do?Thanks again!!

[handesome]

Hi！ Thank you for developing the package！！In your POS_evaluation.R,the return of pos abs(final)，Why did you use absolute values？the pos couldn't be negative?could you please give me an answer？Thanks!

------------------ 原始邮件 ------------------ 发件人: "tinglabs/redPATH" @.>; 发送时间: 2024年2月29日(星期四) 中午12:39 @.>; @.**@.>; 主题: Re: [tinglabs/redPATH] Error in geom_point() (Issue #2)

Hi there!

Thanks for your interest in the package!

For the trajectory plots, could you please check if the variable "trajectory_plots" is empty or does it give you the right plots? I am unsure whether it's the grid.arrange function that might be giving an error here.

The normalizedResultLst is part of the output after running reCAT (tinglabs/reCAT), and it gets saved with this code "save(finalHighCorLst, highCorLst, resultLst, normalizedResultLst, ensembleResultLst, file = paste("bestEnsembleComplexTSP", beginNum, "-", endNum, name, ".RData"))" by default.

After you run get_ordIndex from the reCAT package, you should see a .RData file beginning with "bestEnsembleComplexTSP", and you can load the objects from there.

Best, Ken

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

[kenxie7]

Hi there, sorry I have been traveling.

The POS_score can be negative (ie showing a negative directionality also as shown in the original publication of TSCAN), but I set the absolute because realistically one could determine the start of the differentiation process and I would like to evaluate more on the level of correlation rather than the directionality back then.

Could you please show what you get if you do trajectory_plots[[1]] in R console, if you see plots in the output?

Thanks Ken

[handesome]

Hi!sorry to bother you again.I'm curious about the choice of using bubble sort for calculating the Kendall correlation in your function, rather than using the standard cor function in R. why and which is better?Thanks again!

------------------ 原始邮件 ------------------ 发件人: "han" @.>; 发送时间: 2024年3月25日(星期一) 下午2:55 @.>;

主题: 回复： [tinglabs/redPATH] Error in geom_point() (Issue #2)

Hi！ Thank you for the reply！I‘m truly understand now, and you mean this plot?Thanks!!

------------------ 原始邮件 ------------------ 发件人: "tinglabs/redPATH" @.>; 发送时间: 2024年3月22日(星期五) 晚上10:08 @.>; @.**@.>; 主题: Re: [tinglabs/redPATH] Error in geom_point() (Issue #2)

Hi there, sorry I have been traveling.

The POS_score can be negative (ie showing a negative directionality also as shown in the original publication of TSCAN), but I set the absolute because realistically one could determine the start of the differentiation process and I would like to evaluate more on the level of correlation rather than the directionality back then.

Could you please show what you get if you do trajectory_plots[[1]] in R console, if you see plots in the output?

Thanks Ken

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

[kenxie7]

Hello! The bubble sort index is basically very similar to the Kendall's tau 'distance', as it compares the number of pairwise swaps (ie rank) and computed as a distance (ranges from 0 to 1). However this is not to be confused with the Kendall's tau 'correlation' which is a different correlation coefficient ranging from -1 to 1. So these are two slightly different metrics for assessing the trajectory.