The paper introduces SPAdes, a new genome assembly algorithm designed to address the challenges of single-cell sequencing (SCS) and standard multicell sequencing. SCS faces significant difficulties due to highly non-uniform read coverage, elevated sequencing errors, and chimeric reads. SPAdes improves upon existing assemblers like Velvet and SoapDeNovo, as well as specialized single-cell assemblers like E+V-SC. The key contributions of SPAdes include: 1. **Multisized De Bruijn Graphs**: SPAdes uses multisized de Bruijn graphs to handle non-uniform coverage more effectively than standard de Bruijn graphs. 2. **k-bimer Adjustment**: It introduces k-bimer adjustment to derive accurate distance estimates between k-mers, improving the accuracy of read-pair information. 3. **Paired Assembly Graphs**: Inspired by Paired de Bruijn Graphs (PDBGs), SPAdes constructs paired assembly graphs to better handle variable insert sizes and chimeric reads. 4. **Error Correction**: SPAdes incorporates error correction tools to improve the quality of the final assemblies. The paper outlines the four stages of SPAdes: 1. **Assembly Graph Construction**: Simplifies the de Bruijn graph using multisized graphs, removes bulges, tips, and chimeric reads, and aggregates read-pair information. 2. **k-bimer Adjustment**: Derives accurate distance estimates between k-mers using bireads. 3. **Paired Assembly Graph Construction**: Constructs a paired assembly graph to handle variable insert sizes and chimeric reads. 4. **Contig Construction**: Generates high-quality contigs by backtracking graph simplifications. SPAdes is available online and is distributed as open-source software. The paper also includes benchmarking results showing that SPAdes outperforms other assemblers in both single-cell and multicell datasets.The paper introduces SPAdes, a new genome assembly algorithm designed to address the challenges of single-cell sequencing (SCS) and standard multicell sequencing. SCS faces significant difficulties due to highly non-uniform read coverage, elevated sequencing errors, and chimeric reads. SPAdes improves upon existing assemblers like Velvet and SoapDeNovo, as well as specialized single-cell assemblers like E+V-SC. The key contributions of SPAdes include: 1. **Multisized De Bruijn Graphs**: SPAdes uses multisized de Bruijn graphs to handle non-uniform coverage more effectively than standard de Bruijn graphs. 2. **k-bimer Adjustment**: It introduces k-bimer adjustment to derive accurate distance estimates between k-mers, improving the accuracy of read-pair information. 3. **Paired Assembly Graphs**: Inspired by Paired de Bruijn Graphs (PDBGs), SPAdes constructs paired assembly graphs to better handle variable insert sizes and chimeric reads. 4. **Error Correction**: SPAdes incorporates error correction tools to improve the quality of the final assemblies. The paper outlines the four stages of SPAdes: 1. **Assembly Graph Construction**: Simplifies the de Bruijn graph using multisized graphs, removes bulges, tips, and chimeric reads, and aggregates read-pair information. 2. **k-bimer Adjustment**: Derives accurate distance estimates between k-mers using bireads. 3. **Paired Assembly Graph Construction**: Constructs a paired assembly graph to handle variable insert sizes and chimeric reads. 4. **Contig Construction**: Generates high-quality contigs by backtracking graph simplifications. SPAdes is available online and is distributed as open-source software. The paper also includes benchmarking results showing that SPAdes outperforms other assemblers in both single-cell and multicell datasets.