Dependencies:

JEM-Mapper has the following dependencies:

• MPI library (preferably MPI-3 compatible)
• C++14 (or greater) compliant compiler

Build:

make ksize=$KMER_SIZE

For example: make ksize = 15

Execute:

mpiexec -np $number_of_procs $BINARY -s {Contig_Fasta_File} -q {Long_Read_Fasta_File} -a {A_int_Values_File} -b {B_int_Values_File} -p {Prime_int_Values_File} -r $read_segment length -n $NO_OF_TRIALS

For example:

mpiexec -np 4 ./jem -s ~/Ecoli_reads_100x_contigs.fasta -q ~/Ecoli_reads_10x_long_reads.fasta -a ~/A.txt -b ~/B.txt -p ~/Prime.txt -r 1000 -n 30

Notes:

• This code has been tested on high-performance computing cluster (HPC) with MPI compatibility. For the system we used we had to set the number of processes in the given way. Please change the parameters accordingly.

Input arguments

• -s: input contigs fasta file
• -q: input long reads fasta file
• -a: For diffent trials we have used a linear congruential hash funtion of the form: (Ax+B)%P, whwere A and B are integers and P is a prime and x is a kmer we want to hash. So, using this parameter, we provide the input values for different A values
• -b: input B values
• -p: input prime numbers
• -r: read segment length
• -n: number of trials