PowerNovo2

algorithm · Flow

PowerNovo2: algorithm · Flow. Non-autoregressive generative-flow-based de novo sequencer: successor to PowerNovo v1, designed to skip the cascading-error problem of autoregressive transformers while…

Non-autoregressive generative-flow-based de novo sequencer: successor to PowerNovo v1, designed to skip the cascading-error problem of autoregressive transformers while running 4-5x faster. Published in PLOS Computational Biology, May 2026.

Kind	algorithm
Deep learning	yes
Acquisition	DDA
Family	Flow

Code

• https://github.com/protdb/PowerNovo2

Live stars, open issues and last-push figures are on the Code activity chart.

Data

What the papers describing this method trained it on and reported its results on, as each paper states it. A version is named only where the paper names one.

Evaluated on

• NIST peptide tandem mass spectral libraries (version not stated)
• Nine-species benchmark (version not stated)
• NovoBoard: a comprehensive framework for evaluating the false discovery rate and accuracy of de novo peptide sequencing (as deposited)

Deposited

• PowerNovo2 Comparison data (as deposited)

Paper describing it

• PowerNovo2: A generative flow-based approach to non-autoregressive de novo peptide sequencing (2026, PLOS Computational Biology, peer-reviewed)

Authors (8)

Denis V. Petrovskiy, Kirill S. Nikolsky, Vladimir R. Rudnev, Liudmila I. Kulikova, Tatiana V. Butkova, Kristina A. Malsagova, Arthur T. Kopylov, Anna L. Kaysheva

Seen in the charts

• The architectures
• Code activity
• Models and the authors behind them

Back to the full map