The LHiSA project in C++ is dedicated to users who want to use LHiSA frequently since the implementation (determination of local high-scoring segments and Monte-Carlo simulations) has been optimized to speed its use. It works along with the Templatized C++ Command Line Parser Library (TCLAP - included in the distribution) and the GNU Scientific Library (GSL - required). The executable can be used to perform LHiSA on genetic data or merely to detect local scores in a sequence.

(i) Requirements:

• automake/autoconf
• a C/C++ compiler
• the GNU Scientific Library

(ii) Installation:

• decompress the archive: tar zxvf lhisa-x.x.tar.gz
• compile the package: ./configure then make
• install the program: make install

(iii) Performing LHiSA:

• executable: lhisa

• input data file (required): lhisa -i <filename> or lhisa < <filename> or cat <filename> | lhisa

The input data file must be in the trinary genetic data format. Refer here for further details on this format or download the example.

• output data file (required): lhisa -o <filename> or lhisa > <filename>

• number of Monte-Carlo simulations (optional): lhisa -n <number> It has an impact on the precision of the p-values. The default value (if not set) is 2000 simulations.
• threshold or marker level (optional): lhisa -t <value> Threshold upon which we consider that a marker should contribute positively to the score of segment. This level has an impact on the sensibility and specificity of the method. The default value (if not set) is 10%.

(iv) Detecting local scores segments (loss):

• executable: lhisa -l or lhisa –loss
• input data file (required): lhisa -i <filename> or lhisa < <filename> or cat <filename> | lhisa The input data file is merely a file with the values (in column) of the sequence to submit. Dowload the example.
• output data file (required): lhisa -o <filename> or lhisa > <filename>

(v) Other options:

• help: lhisa -h
• version: lhisa -v

Return to the LHiSA main page