Sleipnir
SeekEvaluator

Usage

Basic Usage

 SeekEvaluator ...

Detailed Usage

package "SeekEvaluator"
version "1.0"
purpose "Evaluate results generated from SeekMiner"

section "Mode"
option  "single"            O   "Evaluate one query's rank list result"
                                flag    off
option  "aggregate"         M   "Evaluate multiple queries and aggregates results"
                                flag    off
option  "multi_weight"      T   "Evaluate multiple queries' dataset weight files"
                                flag    off

section "Metric"
option  "fixed_pr"          z   "Precision at depth X"
                                flag    off
option  "rbp"               r   "Rank biased precision (requires parameter p to be set)"
                                flag    off
option  "avgp"              a   "Average precision for the top X positives, where X = integer, or % of total positives"
                                flag    off
option  "pr"                t   "Precision at X-th positive, where X = integer, or % of total positives"
                                flag    off
option  "pr_all"            c   "Precision at all positive depths (useful for drawing precision-recall curve)"
                                flag    off
option  "auc"               u   "AUC"
                                flag    off
option  "x_int"             x   "Parameter X = integer, for --avgp, --pr"
                                int default="-1"
option  "x_per"             e   "Parameter X = percentage, for --avgp, --pr"
                                float   default="0"
option  "rbp_p"             p   "Parameter p, for --rbp"
                                float   default="0.95"
option  "dislay_only"       F   "Display the genes sorted by score (top 500 is shown, for single mode only)"
                                flag    off
option  "display_weight"    W   "Display dataset weights (top 100)"
                                flag    off

section "Aggregator (for multi-query evaluation)"
option  "agg_avg"           A   "Show the average, standard deviation of the metric for all queries"
                                flag    off
option  "agg_quartile"      B   "Show the min, max, as well as the 1st, 2nd, 3rd quartile of the metric for all queries"
                                flag    off
option  "agg_ranksum"       C   "Sum up the ranks of genes in all query rankings to produce a master list sorted by summed rank, and perform metric on this list"
                                flag    off
option  "agg_scoresum"      D   "Sum up the scores of genes in all query rankings to produce a master list sorted by summed score, and perform metric on this list"
                                flag    off
option  "display_all"       E   "Display the metric for all queries"
                                flag    off

section "Input required by all"
option  "input"             i   "Gene mapping file"
                                string typestr="filename"   yes

section "Input required by dataset weight display (single .dweight file)"
option  "dataset_map"       I   "Dataset mapping file, only required for displaying dataset weights"
                                string typestr="filename"
option  "weight"            w   "Dataset weight file, (*.dweight)"
                                string typestr="filename"

section "Input required by multi-query dataset weight display"
option  "dweight_list"      Z   "List of dweight files"
                                string typestr="filename"

section "Options for all"
option  "fold_over_random"  f   "Fold-over-random"
                                flag    off
option  "p_value"           P   "Simulated p-value"
                                flag    off
option  "random_dir"        R   "Random directory"
                                string typestr="directory"
option  "random_num"        N   "Number of random trials"
                                int     default="100"

section "Input required by single-query"
option  "goldstd"           s   "Gold standard gene set file (one line, space delimited)"
                                string typestr="filename"
option  "gscore"            g   "Gene score file (.gscore)"
                                string typestr="filename"
option  "query"             q   "Query gene set file (to be excluded from evaluation) (.query)"
                                string typestr="filename"
option  "exclude"           y   "Exclude genes (.exclude)"
                                string typestr="filename"
option  "nan"               n   "Define NaN score"
                                float   default="-320"
                                
section "Input required by multi-query"
option  "goldstd_list"      S   "List of gold standard gene set files"
                                string typestr="filename"
option  "gscore_list"       G   "List of gene score files"
                                string typestr="filename"
option  "query_list"        Q   "List of query gene set files"
                                string typestr="filename"
option  "exclude_list"      X   "Exclude gene list"
                                string typestr="filename"
option  "include_list"      Y   "List of annotated genes"
                                string typestr="filename"


section "Output"
option  "dir_out"           d   "Output directory"
                                string typestr="directory"  yes

Flag Default Type Description
-i stdin Text file Tab-delimited text file containing two columns, numerical gene IDs (one-based) and unique gene names (matching those in the input DAT/DAB files).
-d . Directory Input directory containing DB files
-D . Directory Output directory in which database files will be stored.
-x . Text file Input file containing list of CDatabaselets to combine