Update quickstart.rst

marbl · Jul 13, 2016 · 004c999 · 004c999
1 parent f55f64a
commit 004c999
Showing 1 changed file with 52 additions and 21 deletions.
diff --git a/docs/source/quickstart.rst b/docs/source/quickstart.rst
@@ -73,24 +73,55 @@ Options
 -----------------
 The full list of options is available via command-line help (--help or -h). Below is a list of commonly used options.
 
-   -h                  Displays the help menu.
-   --version           Displays the version.
-   --settings          Set all unset parameters for the default settings. Same defaults are applied to Nanopore and Pacbio reads. 0) None, 1) Default, 2) Fast, 3) Sensitive.
-   -k                		[int], k-mer size used for MinHashing. The k-mer size for second stage filter is seperate, and cannot be modified, default = 16
-   -q                		Usage 1: The FASTA file of reads, or a directory of files, that will be compared to the set of reads in the box (see -s). Usage 2: The output directory for the binary formatted dat files, default="".
-   -s                		Usage 1 only. The FASTA or binary dat file (see Usage 2) of reads that will be stored in a box, and that all subsequent reads will be compared to, default = ""
-   -p                		Usage 2 only. The directory containing FASTA files that should be converted to binary format for storage, default = ""
-   --num-hashes        [int], number of min-mers to be used in MinHashing, default = 512.
-   --num-min-matches   [int], minimum # min-mer that must be shared before computing second stage filter. Any sequences below that value are considered non-overlapping, default = 3.
-   --threshold         [double], the threshold cutoff for the second stage sort-merge filter. This is based on the mash distance computed from the Jaccard distance of k-mers (size given by ordered-kmer-size) in the overlapping regions, default = 0.78
-   --max-shift         [double], region size to the left and right of the estimated overlap, as derived from the median shift and sequence length, where a k-mer matches are still considered valid. Second stage filter only, default = 0.2.
-   --ordered-kmer-size  [int], The size of k-mers used in the ordered second stage filter, , default = 12.
-   --ordered-sketch-size  [int], The sketch size for second stage filter, default = 1536.
-   --filter-threshold  [double], the cutoff at which the k-mer in the k-mer filter file is considered repetitive. This value for a specific k-mer is specified in the second column in the filter file. If no filter file is provided, this option is ignored, default = 1.0E-5.
-   --repeat-weight        [double] Repeat suppression strength for tf-idf weighing. <0.0 do unweighted MinHash (version 1.0), >=1.0 do only the tf weighing. To perform no idf weighting, do no supply -f option, default = 0.0
-			--supress-noise 			    [int] 0) Does nothing, 1) completely removes any k-mers not specified in the filter file, 2) supresses k-mers not specified in the filter file, similar to repeats, default = 0
-   -f                     [string], k-mer filter file used for filtering out highly repetative k-mers. Must be sorted in descending order of frequency (second column), default = "".
-   --no-self           Do not compute the overlaps between sequences inside a box. Should be used when the to and from sequences are coming from different files, default = false.
-   --no-tf             Do not perform the tf weighing, of the tf-idf weighing, default = false
-			--num-threads       [int], number of threads to use for computation. Typically set to #cores, , default = 4.
-			--store-full-id        Store full IDs as seen in FASTA file, rather than storing just the sequence position in the file. Some FASTA files have long IDS, slowing output of results. This options is ignored when using compressed file format, default = false.
+	Usage 1 (direct execution): java -server -Xmx<memory> -jar <MHAP jar> -s<fasta/dat from/self file> [-q<fasta/dat to file>] [-f<kmer filter list, must be sorted>]
+
+	Usage 2 (generate precomputed binaries): java -server -Xmx<memory> -jar <MHAP jar> -p<directory of fasta files> -q <output directory> [-f<kmer filter list, must be sorted>]
+
+		--filter-threshold, default = 1.0E-5
+			[double], the cutoff at which the k-mer in the k-mer filter file is considered repetitive. This value for a specific k-mer is specified in the second column in the filter file. If no filter file is provided, this option is ignored.
+		--help, default = false
+			Displays the help menu.
+		--max-shift, default = 0.2
+			[double], region size to the left and right of the estimated overlap, as derived from the median shift and sequence length, where a k-mer matches are still considered valid. Second stage filter only.
+		--min-olap-length, default = 116
+			[int], The minimum length of the read that used for overlapping. Used to filter out short reads from FASTA file.
+		--min-store-length, default = 0
+			[int], The minimum length of the read that is stored in the box. Used to filter out short reads from FASTA file.
+		--no-self, default = false
+			Do not compute the overlaps between sequences inside a box. Should be used when the to and from sequences are coming from different files.
+		--no-tf, default = false
+			Do not perform the tf weighing, in the tf-idf weighing.
+		--num-hashes, default = 512
+			[int], number of min-mers to be used in MinHashing.
+		--num-min-matches, default = 3
+			[int], minimum # min-mer that must be shared before computing second stage filter. Any sequences below that value are considered non-overlapping.
+		--num-threads, default = 8
+			[int], number of threads to use for computation. Typically set to #cores.
+		--ordered-kmer-size, default = 12
+			[int] The size of k-mers used in the ordered second stage filter.
+		--ordered-sketch-size, default = 1536
+			[int] The sketch size for second stage filter.
+		--repeat-weight, default = 0.9
+			[double] Repeat suppression strength for tf-idf weighing. <0.0 do unweighted MinHash (version 1.0), >=1.0 do only the tf weighing. To perform no idf weighting, do no supply -f option. 
+		--settings, default = 0
+			Set all unset parameters for the default settings. Same defaults are applied to Nanopore and Pacbio reads. 0) None, 1) Default, 2) Fast, 3) Sensitive.
+		--store-full-id, default = false
+			Store full IDs as seen in FASTA file, rather than storing just the sequence position in the file. Some FASTA files have long IDS, slowing output of results. This options is ignored when using compressed file format.
+		--supress-noise, default = 0
+			[int] 0) Does nothing, 1) completely removes any k-mers not specified in the filter file, 2) supresses k-mers not specified in the filter file, similar to repeats. 
+		--threshold, default = 0.78
+			[double], the threshold cutoff for the second stage sort-merge filter. This is based on the identity score computed from the Jaccard distance of k-mers (size given by ordered-kmer-size) in the overlapping regions.
+		--version, default = false
+			Displays the version and build time.
+		-f, default = ""
+			k-mer filter file used for filtering out highly repetative k-mers. Must be sorted in descending order of frequency (second column).
+		-h, default = false
+			Displays the help menu.
+		-k, default = 16
+			[int], k-mer size used for MinHashing. The k-mer size for second stage filter is seperate, and cannot be modified.
+		-p, default = ""
+			Usage 2 only. The directory containing FASTA files that should be converted to binary format for storage.
+		-q, default = ""
+			Usage 1: The FASTA file of reads, or a directory of files, that will be compared to the set of reads in the box (see -s). Usage 2: The output directory for the binary formatted dat files.
+		-s, default = ""
+			Usage 1 only. The FASTA or binary dat file (see Usage 2) of reads that will be stored in a box, and that all subsequent reads will be compared to.