search

vahuynh / dynGENIE3

Semi-parametric approach for the inference of gene regulatory networks from time series of expression data
21 stars 17 forks source link

Error in if (n > 0) c(NA_integer_, -n) else integer() #8

Open JiakaiLiang opened 8 months ago

JiakaiLiang commented 8 months ago

Hi,

I encountered an error while attempting to utilize all quantified genes in order to construct a comprehensive regulatory network, from which I aim to extract subnetworks containing the genes of interest. The transcriptome data was quantified using triploid typing genomes, encompassing 6 time points with 3 replicates each (totaling 18 samples). For file designation, replicates 1 of each time point were labeled as "time_series_1," replicates 2 as "time_series_2," and replicates 3 as "time_series_3." To address the issue of numerous genes exhibiting expression values of zero or partial samples with zero expression values, only those genes displaying expression levels exceeding 0.1 TPM across all samples were utilized for program execution.

Error in if (n > 0) c(NAinteger, -n) else integer() : missing value where TRUE/FALSE needed Calls: get.link.list ... melt -> melt.matrix -> expand.grid -> .set_row_names In addition: Warning messages: 1: In rep.fac * nx : NAs produced by integer overflow 2: In .set_row_names(as.integer(prod(d))) : NAs introduced by coercion to integer range Execution halted

I sincerely look forward to your assistance ！

JiakaiLiang commented 8 months ago

Additionally, I would appreciate clarification on the correctness of the method employed in generating TS files. If it is deemed incorrect, I would greatly appreciate any advice or suggestions.

vahuynh commented 7 months ago

Hi,

The errors seems to occur when you call the get.link.list function. Did you check if you have any NAs in the data (that you used as input to dynGENIE3) or in the weight matrix returned by dynGENIE3?