Input Files
Parameters file
An optional parameters file can be used in an ATHENA run. The format is keyword: value and parameters specified in the file take precedence over ones on the command line. The parameters file is a useful way to replicate previous runs of ATHENA. See the Parameters list for all available options.
GRAMMAR_FILE: example/genn.bnf
OUTCOME_FILE: example/outcome_contin.txt
CONTIN_FILE: example/contin.txt
SCALE_CONTIN: True
SCALE_OUTCOME: True
GENO_FILE: example/genos.txt
GENO_ENCODE: add_quad
OUT: results/athena
FITNESS: r-squared
POP_SIZE: 250
GENERATIONS: 5
Data files
ATHENA data files are whitespace delimited and the first row must contain the column headers. A data file must have an ID column followed by the labels for the data (genotypic, continuous, or outcome) included in the file. The ID column is used to match the data presented in rows across the various input files.
An outcome file and at least one of the genotype and continuous data files are required for an ATHENA run.
Genotype file
A genotype input file must contain the ID column. The genotypes must be 0,1,2 or the value designated by the MISSING parameter
ID SNP1 SNP2 SNP3 SNP4
id1 1 2 0 0
id2 0 2 0 0
id3 1 1 1 0
id4 0 0 1 1
id5 0 1 1 0
id6 0 1 1 1
id7 0 0 1 1
Continuous data file
A continuous data file must contain the ID column. Data can be any numeric values or the value designated by the MISSING parameter
ID C1 C2 C3 C4
id1 -0.19824033325 -0.25884851921 1.29742330517 0.04194009721
id2 -0.32127593922 0.04965043519 1.52750412357 0.04550146096
id3 -0.89618956836 -0.2082694151 0.83338920376 -0.12484128176
id4 -0.00847565279 0.15470277072 0.26226553399 -0.00364302721
id5 -0.52438207799 0.19822901217 1.19583036534 -0.04283847354
id6 0.23387312884 -0.11317690779 0.64984994477 -0.07093586693
id7 -0.23458885217 0.27120910755 1.18926694272 -0.10301740165
Outcome file
The outcome file is required and must contain an ID column
ID outcome1 outcome2
id1 0.4056314684 0.7949572102
id2 -0.7938889637 -0.7737498065
id3 1.291416007 -0.0263485549
id4 -0.241405327 1.064138677
id5 -0.1028663478 0.0355249568
id6 0.1577496992 -0.0535040572
id7 0.5924636053 -0.8850819371
Testing files
The testing files are designated by parameters TEST_OUTCOME_FILE, TEST_GENO_FILE and TEST_CONTIN_FILE file. The formats match the equivalent input files described above. The testing files are used in a run where training is done on the data files provided by OUTCOME_FILE, GENO_FILE and CONTIN_FILE and the testing is done on these files. In other words, it allows the user to split the data before a run.
Grammar file
The grammar file defines the collection of terminal symbols, non-terminal symbols, production rules and the start symbol for creating networks. ATHENA automatically adjusts the production rule for the
<p> ::= <pn>(<pinput>)
<pn> ::= PA
| PS
| PM
| PD
<pinput> ::= [<winput>,<winput>]
| [<winput>,<winput>,<winput>]
| [<winput>,<winput>,<winput>,<winput>]
| [<winput>,<winput>,<winput>,<winput>,<winput>]
<winput> ::= (<cop> * <v>)
| (<cop> * <p>)
<cop> ::= float(<num>)
| <e>
<e> ::= (<cop> + <cop>)
| (<cop> - <cop>)
| (<cop> * <cop>)
| pdiv(<cop>,<cop>)
<num> ::= <dig>.<dig>
| .<dig>
| .<dig><dig>
| <dig>.<dig><dig>
| <dig><dig>.<dig><dig>
<dig> ::= 0
| 1
| 2
| 3
| 4
| 5
| 6
| 7
| 8
| 9
<v> ::= x[0] | x[1] | x[2] | x[3] | x[4] | x[5] | x[6] | x[7] | x[8] | x[9]
Color mapping file
The color mapping file specifies colors for specific variables in the input files. If the variables appear in the networks plotted at the end of the run, those network nodes will appear in the designated colors.
The file has 3 columns that are tab-delimited: Category, Color, Inputs. The Color column accepts X11/CSS colors or hex colors (#ff5733)
Category Color Inputs
APOE orange SNP3 SNP92 SNP22
ABCA7 lightblue SNP12 SNP27
Environmental lightgreen C30 C2
Response to cytokine royalblue SNP79