iMapSplice

iMapSplice is an individualized RNA-seq read alignment software. It makes use of personal genomic information and performs an unbiased alignment towards a genome index carrying both the reference and any alternative bases. It is light-weight and does not require an index to be rebuilt for each individual. Importantly, it breaks the computational emphasis or dependency on the position of canonical splice site dinucleotide motifs in the reference genome and enables iMapSplice to discover personal splice junctions created through splice site mutations.

Installation

System Requirements

iMapSplice has been tested on Linux platforms with the following system settings.

• Red Hat 4.4.5-6
• gcc version 4.9.1

Installation

1. Clone the repository:

   git clone https://github.com/xa6xa6/iMapSplice.git
   

2. Build and install:

   cd code
   make all
   

   After successfully build the code, you can find iMapSplice toolchain at bin/.

   ls bin/
   

Manual

1. Building reference genome index (MapSplice index, only need to run once, shared by all the individuals)

   Note: before building index, you need to put all the sequence files of reference genome into a directory (like /PATH/hg19/). And all the sequence files are required to be in the following format:

   (1) In "FASTA" format, with ".fa" extension. (2) One chromosome per sequence file. (3) Chromosome name in the header line (">" not included) is the same as the sequence file base name, and does not contain any blank space. E.g. If the header line is ">chr1", then the sequence file name should be "chr1.fa". (5) No other files in the same folder.

   ./buildWholeGenome <input_chromosomes_folder_path> <output_globalIndex_folder_path>
   ./build2ndLevelIndex <input_globalIndex_folder_path> <output_localIndex_folder_path>
   

2. Generating SNP-mers

   Generating SNP-mers with phased SNPs

   ./getSNPmer-phased <input_globalIndex_folder_path> <input_GAF_path> <input_SNPlist_path(phased)> <output_SNPmer_folder_path> <SNPmer_length>
   

   or

   Generating SNP-mers with unphased SNPs

   ./getSNPmer-unphased <input_globalIndex_folder_path> <input_GAF_path> <input_SNPlist_path(unphased)> <output_SNPmer_folder_path> <SNPmer_length> <Sampling_num_max>
   

   Note: We suggest using 201 for SNPmer_length and 5 for sampleing_num_max for human genome and other genomes in similar size. If you are working with a much smaller genome, considering using shorter SNP-mers, eg. 51-mer or 31-mer.

   Note: SNP list (tab delimited, four columns):

   ...
   chr1 1000 A T
   chr2 2000 C G
   chrX 3000 G T 
   ...
   

   Each row records a SNP: Column 1: chromosome name Column 2: chromosome pos Column 3: Haplotype 1 base (phased) or reference base (unphased) Column 4: Haplotype 2 base (phased) or alternate base (unphased)

3. Building SNP-mer index

   ./buildSNPmerIndex <input_SNPmer_folder_path/SNPinAnn.fa> <output_SNPmerIndex_folder_path> 
   

4. Mapping

   Mapping with phased SNPs

   ./iMapSplice-phased -P <input_SNPlist_path(phased)> -Q <input_SNPmerIndex_folder_path> -G <input_globalIndex_folder_path> -L <input_localIndex_folder_path> -1 <read_end1> -2 <read_end2> -T <threads_num> -O <output_folder>
   

   or

   Mapping with unphased SNPs

   ./iMapSplice-unphased -P <input_SNPlist_path(unphased)> -Q <input_SNPmerIndex_folder_path> -G <input_globalIndex_folder_path> -L <input_localIndex_folder_path> -1 <read_end1> -2 <read_end2> -T <threads_num> -O <output_folder>
   

License

Please refer to LICENSE.txt