The paper introduces PEAR, a new software tool designed for merging Illumina paired-end reads from target fragments of varying lengths. PEAR evaluates all possible paired-end read overlaps and does not require the target fragment size as input. It implements a statistical test to minimize false-positive results. Tests on simulated and empirical data show that PEAR consistently generates highly accurate merged paired-end reads. The program is optimized for efficiency, allowing it to merge millions of paired-end reads within minutes on standard desktop computers and achieving linear speedups on multi-core architectures. PEAR is implemented in C and uses POSIX threads, and it is freely available at http://www.exelixis-lab.org/web/software/pear. The authors compare PEAR with other state-of-the-art mergers (FLASH, PANDAseq, and COPE) and demonstrate its superior performance, especially in scenarios where other tools fail due to varying fragment lengths.The paper introduces PEAR, a new software tool designed for merging Illumina paired-end reads from target fragments of varying lengths. PEAR evaluates all possible paired-end read overlaps and does not require the target fragment size as input. It implements a statistical test to minimize false-positive results. Tests on simulated and empirical data show that PEAR consistently generates highly accurate merged paired-end reads. The program is optimized for efficiency, allowing it to merge millions of paired-end reads within minutes on standard desktop computers and achieving linear speedups on multi-core architectures. PEAR is implemented in C and uses POSIX threads, and it is freely available at http://www.exelixis-lab.org/web/software/pear. The authors compare PEAR with other state-of-the-art mergers (FLASH, PANDAseq, and COPE) and demonstrate its superior performance, especially in scenarios where other tools fail due to varying fragment lengths.