Cannot guess the genome build, please set it by hand!

[jingydz]

command

vcf_files <- c("test1.snp.vcf","test2.snp.vcf", "test3.snp.vcf") maf <- read_vcf(vcf_files) mt_tally <- sig_tally( maf, ref_genome = "BSgenome.Sscrofa.UCSC.susScr11", useSyn = TRUE )

logfile

i [2022-04-15 08:50:12]: Started. i [2022-04-15 08:50:13]: We would assume you marked all variants' position in + strand. x [2022-04-15 08:50:13]: Cannot guess the genome build, please set it by hand! i [2022-04-15 08:50:13]: 1.165 secs elapsed.

How can I to build the genome index?

[ShixiangWang]

@jingydz May you try:

mt_tally <- sig_tally(
maf,
ref_genome = "BSgenome.Sscrofa.UCSC.susScr11",
genome_build = "susScr11",
useSyn = TRUE
)

You must set genome_build to a non NULL value here, otherwise the code infer the genome_build from your ref_genome. Please also note that due to no annotation data available in sigminer for susScr11, you must not set add_trans_bias to TRUE (just keep it as default).

[jingydz]

thanks! It worked but in the next steps, I cann't run it successfully. It paused for a long time when I run the next command, maybe it's because I use my PC to run the pipeline.

command

mt_est <- sig_estimate(mt_tally$nmf_matrix, range = 2:20, nrun = 10, # increase this value if you wana a more stable estimation use_random = FALSE, # if TRUE, add results from randomized input cores = 1,

pConstant = 1e-13,

                    verbose = TRUE)

[ShixiangWang]

Yeah. In you PC, you can try range 2:5 and nrun=3 to see if it works. If your matrix, nrun and range are large, you should take a linux server to run the task.

[ShixiangWang]

Hi, I am closing this now. Please file another issue if you have any problem.