The "LiBiNorm model" mode takes the landscape file that can be generated by LiBiNorm count using the -u <filenameroot> and -l option and provides further bias analysis options. This avoids the need to reprocess the bam and feature files when investigating bias characteristics for a specific set of data.
The command line format is LiBiNorm model [options] <landscape file>
The -l option in "LiBiNorm count" produces a landscape file that contains the locations of the reads in each of the genes/transcripts. There are additional options available for analysing the bias using the "LiBiNorm model" mode which loads in the previously created landscape file. The "LiBiNorm count" -n, -u, -p and -d options are available as before. In addition:
- The number of Markov Chain Monte Carlo (MCMC) runs can be changed from the default of 10 using -r N.
- The length of each MCMC run can be changed from the default of 200 using -s N.
- The normal mode is that parameters are calculated for all models. The -m mode specifies that only one model is run with the default number of runs. A single mcmc run is performed for all other models.
- A limit can be set on the number of genes that are used for the parameter estimation using the -g option.
- The -f option provides more extensive TPM and FPKM data in the output, with and without bias normalisation.
The process of parameter determination uses the Nelder Mead simplex method to locate an initial set of parameters for the second stage which uses MCMC to identify the uncertainty associated with each of the parameters. The -o option results in the initial Nelder Mead stage being omitted and only the MCMC stage being used to determine the coefficients. When this is done, the parameters for the starting point for each run are determined by a random selection from within the likely range of parameters which is used within LiBiNorm as a prior. When this mode is selected then it is necessary to use more (-r), and longer (-s) runs.