SemRep.v1.8_Options.txt

*Title: SemRep Options
*Author: Halil Kilicoglu

*TOC 

* Introduction

SemRep can be run on the command line, the default input and output are
standard input and output. SemRep also allows specifying input and output
files on the command line as follows:

^<<
% semrep.v1.8 [Options] InputFile OutputFile
^>>

The InputFile and OutputFile arguments, if specified, must be the last two
arguments. It is not necessary to specify OutputFile, because it will default
to <InputFile>.sem.<Version> (<Version> in this case is v.1.8). Note that if 
the output file is an existing file, it will be overwritten.

NOTE: As of SemRep 1.8, some MetaMap options can be directly invoked from
SemRep (e.g., -J (--restrict_to_sts) or -e (--exclude_sources)). For more
information on MetaMap options, see [MetaMap Options (https://metamap.nlm.nih.gov/Docs/MM_2016_Usage.pdf)].
For a list of currently available SemRep options, use -h (--help) option. 

* Data Options

-A (--anaphora_resolution): performs sortal anaphora resolution (expressions such as 'this disease',
'these agents', etc.). 

-D (--dysonym_processing): blocks MetaMap from using certain UMLS synonyms that are 
deemed to be harmful (dysonyms) in mapping to concepts (e.g., 'prostate' is a dysonym
for 'prostate cancer'). Used by default.

-S (--generic_processing): standard (non-domain-specific) SemRep processing. Used by default.

-L (--lexicon_year <year>): allows specifying a UMLS SPECIALIST Lexicon release. Currently,
available options are '2006' and '2018'. It defaults to '2018'.

-M (--relaxed_model): allows running MetaMap in relaxed mode. 

-n (--use_generic_domain_modification): allows the use of a hand-crafted, supplementary
knowledgebase that 'corrects' some of the semantic type information in UMLS (e.g.,
'Smoker' (fndg) -> 'Smoker' (popg,humn) and adds new synonyms for some existing concepts
(e.g., 'loci' for 'genetic locus' (gngm)).

-N (--use_generic_domain_extensions): equivalent to using -n option, in addition to defining
new concepts that are absent in UMLS (e.g., 'cancer survival' (orgf) with the synonym 'cancer survival').

-V (--mm_data_version <version>): allows specifying the UMLS data version to use. 
'USAbase' version is available with the current distribution. In case another version is desired,
the relevant files must be downloaded first from [MetaMap Datasets (https://metamap.nlm.nih.gov/MetaMap_Optional_Datasets.shtml)] and installed to MetaMap DB directory. 

-y (--word_sense_disambiguation): allows using MetaMap's word sense disambiguation module for
UMLS concepts. Ued by default.

-Z (--mm_data_year <release>): allows specifying the UMLS data release to use. By default,
SemRep uses '2006AA'. The other available release in this version is '2018AA'.

* Output Options

-F (--full_fielded_output_format): generates a more verbose output than the standard 
output. Details for this type of output can be found in [Full Fielded Output documentation 
(https://semrep.nlm.nih.gov/semrep_full_fielded_output.pdf)].

-P (--extract_phrases_only): generates output that consists of various types of 
noun phrases (simple, mega, macro). For more details, see [PhraseX (https://ii.nlm.nih.gov/MTI/phrasex.shtml)].

-R (--write_syntax): generates a readable output of the minimal commitment parser.

-r (--write_syntax_only): generates only the output of the minimal commitment parser.

-U (--expanded_utterances_only): outputs only sentences with the acronyms expanded.

-u (--unexpanded_utterances_only): outputs only sentences with no acronym expansion performed.

-X (--xml_output_format): generates verbose XML output, the content of which is similar to full fielded format. 
For more details, see [XML output documentation (https://semrep.nlm.nih.gov/SemRep.v1.8_XML_output_desc.html)]

Note: With no options specified, semrep.v1.8 shell script will invoke the following options: 
-DSy -L 2006 -Z 2006AA --negex_st_set ALL. If you need to change this configuration, you may edit
semrep.v1.8 script or create another script based on semrep.v1.8.