Issue with simpleLoess

[asan-emirsaleh]

Hello!
I am enjoyed on the abilities the sleuth provides on RNA-Seq data analysis. So I try to use it in my current project. I have a large dataset, where from I am interested in looking the DE for 15 genes (which corresponds to 224 DE transcripts previously identified by lrt analysis. I ran an analysis again, with gene aggregation, target mapping and filtering by a vector containing the 224 transcript ids.

In sleuth_prep(s2c, num_cores = max(1L, parallel::detectCores() -  :
  325270 target_ids are missing annotations for the aggregation_column: gene.
These target_ids will be dropped from the gene-level analysis.
If you did not expect this, check your 'target_mapping' table for missing values.
fitting measurement error models
shrinkage estimation
Error in simpleLoess(y, x, w, span, degree = degree, parametric = parametric,  : 
  invalid 'x'
Calls: sleuth_fit -> shrink_df -> eval -> loess -> simpleLoess
Execution halted

The commands used:

t2g <- read.csv('DEG_vs_comb_prot_dedup.map.tsv', sep="\t", stringsAsFactors = FALSE)
head(t2g)
filter_id <- t2g$target_id[t2g$gene != '']

so <- sleuth_prep(s2c, 
   num_cores = max(1L, parallel::detectCores() - 10L), 
   extra_bootstrap_summary = TRUE, 
   read_bootstrap_tpm = TRUE, 
   target_mapping = t2g, 
   aggregation_column = "gene", 
   gene_mode = TRUE, 
   filter_target_id = filter_id, 
   max_bootstrap = 500)

[apsteinberg]

Were you able to resolve this in the end? I am getting a similar error.