Performance evaluation of six popular short-read simulators.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• High-throughput sequencing generates vast genomic data.
• Interpreting genomic data requires understanding computational analysis methods.
• Synthetic sequencing data can assess computational pipelines for diverse species and studies.

Purpose of the Study:

• To evaluate the performance of six popular short-read simulators.
• To understand the ability of read simulators to emulate empirical genomic data characteristics.
• To provide guidance on selecting appropriate simulators for benchmarking computational pipelines.

Main Methods:

• Comparison of six short-read simulator software: ART, DWGSIM, InSilicoSeq, Mason, NEAT, and wgsim.
• Analysis of simulator performance in emulating genomic characteristics.
• Discussion of considerations for selecting suitable simulation models.

Main Results:

• The performance of the evaluated short-read simulators in emulating empirical genomic data varies significantly.
• Understanding these variations is crucial for accurate computational pipeline benchmarking.
• No single simulator is universally optimal for all benchmarking needs.

Conclusions:

• The choice of short-read simulator impacts the reliability of computational pipeline evaluations.
• Further research is needed to improve the fidelity of read simulators.
• Careful selection of simulators is essential for robust genomic data analysis and interpretation.