NMhmmfit

We assume here that, as described in the previous section, you have created for each strain a data file in NucleoMiner format using NMtdb, so that binary file Data/BY_S288c/Array/NucOccupancy_norm.db contains the average signal among (biological) replicates from the nucleosome mapping chip experiment.

For speed consideration, and since the process is the same for each chromosome, we will fit here the HMM only on chromosome I of BY_S288c (using the -c option of NMhmmfit).

$NMhmmfit -i Data/BY_S288c/Array/NucOccupancy_norm.db \
          -o S2/BY_S288c/NMhmmfit -c chrI
>chrI|nprobe=46546|nwindow=57515|coverage=0.81|
gdl_tiling_hmm_baum_welch stopped: the tiling sequence cannot be built from the HMM!
gdl_tiling_hmm_baum_welch stopped: the tiling sequence cannot be built from the HMM!
gdl_tiling_hmm_baum_welch stopped: the tiling sequence cannot be built from the HMM!
gdl_tiling_hmm_baum_welch stopped: the tiling sequence cannot be built from the HMM!
gdl_tiling_hmm_baum_welch stopped: the tiling sequence cannot be built from the HMM!
0%|=          | 10%

As you can see, sometimes the message

gdl_tiling_hmm_baum_welch stopped: the tiling sequence cannot be built from the HMM!

appears. Don't worry too much about this message: it is just a warning. It means that for a given sliding window, the program was unable to fit the HMM to the data within that window. For example, this can happen when the window contain a large amount of missing data, or the data quality, or more precisely the signal to noise ratio is low.

Now, as you can see, the program NMhmmfit has created a new directory S2/BY_S288c/NMhmmfit/ which contains a new file:

$ls S2/BY_S288c/NMhmmfit/
chrI.nhw

Don't try to open this file with a text editor. Again this is a binary file,

$file S2/BY_S288c/NMhmmfit/chrI.nhw
S2/BY_S288c/NMhmmfit/chrI.nhw: data

and you must not modify this file, otherwise you will experiment troubles when using NMhmmvit.