Scroll to navigation



kisSplice - local assembly of SNPs, indels and AS events


kissplice -h
kissplice [OPTION] [-r READFILES]


Detects alternative splicing events and other kinds of polymorphisms from READFILES (in FASTA or FASTQ format).


Show this help message and exit.
Input fasta/q read files or compressed (.gz) fasta/q files (mutiple, such as "-r file1 -r file2...")
k-mer size (default=41).
Maximum number of branching nodes (default: 5)
Maximal length of the shorter path (default: 2k+1).
Minimum length of the shorter path (default 2k-8).
Maximum length of the longest path (default: 1000000), skipped exons longer than UL_MAX are not reported.
Path and prefix to pre-built de Bruijn graph (suffixed by .edges/.nodes) if jointly used with -r, graph used to find bubbles and reads used for quantification.
Path to store the results (default = ./results).
Specific directory (absolute path) where to build temporary files (default temporary directory otherwise).
Number of cores (must be <= number of physical cores).
0, 1 or 2. Changes which types of SNPs will be output. If 0 (default), will not output SNPs. If 1, will output Type0a-SNPs. If 2, will output Type0a and Type0b SNPs (warning: this option may increase a lot the running time.).
Verbose mode.
Keep the nodes/edges file for unfinished bccs.
Discard k-mers tha are present strictly less than this number of times in the dataset. (default 2).
Discard edges with relative coverage below MIN_RELATIVE_COV expressed as a percentage in [0,1). (default 0.05).
Classify as inexact repeats those bubbles whose paths' edit distance is smaller than MIN_EDIT_DIST (default 3).
Maximal number of bubble enumerations in each bcc. If exceeded, no bubble is output for the bcc (default: 100000000).
Maximal number of substitutions authorized between a read and a fragment (for quantification only), default 2. If you increase the mismatch and use --counts think of increasing min_overlap too.
Changes how the counts will be reported. If 0: total counts, if 1: counts on junctions, if 2 (default): all counts.
Sets how many nt must overlap a junction to be counted by --counts option (default: 5).
Max amount of time (in seconds) spent for enumerating bubbles in each bcc. If exceeded, no bubble is output for the bcc (default 100000).
Display program's version number and exit.
Will output the maximum non-ambiguous context of a bubble.
Will output the id of the nodes composing the two paths of the bubbles.
Will output the number of branching nodes in each path.
Keep the node/edges files for all bccs.
Do not use a new experimental algorithm that searches for bubbles by listing all paths.
If you use the experimental algorithm, you must provide the maximum size of the process's virtual memory (address space) in megabytes (default unlimited).
Keep the .counts file after the sequencing-errors-removal step.
Creates a file with the KissReads mapping information of the reads on the bubbles.
Execute kissreads in stranded mode.
Sets the minimum number of reads mapping to a path of a bubble in a read set is needed to call a strand.
If a strand is called for a path of a bubble in a read set, but the proportion of reads calling this strand is less than this threshold, then the strand of the path is set to '?' (any strand - not enough evidence to call a strand).
Keep the Type_1 redundant cycles in the result file.
Keep the low-complexity Type_1 cycles in the result file.
Cycles with a Shannon entropy value for their upper path below this value will be removed (use --keep-low-complexity to keep them).
Creates files with informations on compressed redundant cycles.
Creates a file with informations on removed low-complexity cycles.
Only quantify Type 1 bubbles (alternative splicing events, MAJOR SPEED UP with -b > 10 BUT all other bubbles will not appear in the result file).
2024-01-09 kissplice 2.6.2