Fig. 1

Simplified workflow of the haplotype assembly process. Raw sequencing data are initially aligned, defining m reads. Every position of the two chromosome copies is compared against a reference chromosome. The black solid points denote n heterozygous positions, along with the corresponding nucleobases. The fragment matrix M is defined assigning 1 to SNP positions and 0 to wild-type positions. To reconstruct the two haplotypes h₁ and h₂ characterized by the least number of corrections to the SNP values among the 2ⁿ candidate haplotypes, the wMEC problem is solved by partitioning the matrix M into two disjoint matrices M₁ and M₂