BiATNovo

algorithm · Transformer (AR)

BiATNovo: algorithm · Transformer (AR). Bidirectional self-attention

Bidirectional self-attention

Kind	algorithm
Deep learning	yes
Acquisition	both
Family	Transformer (AR)

Code

https://github.com/yangshu729/biatNovo-DDA

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

Benchmarks

denovo_benchmarks: median peptide-level average precision 0.469 over 86 datasets, median rank 11 of 14 (version bm-1.0.0).

Both are mass-based matches on the tool’s most recent run. What these numbers mean.

Reported comparisons (2)

The comparison tables this method’s own papers print, standardised: every value on a 0-1 scale, methods down the side, the measure and then the species across. These are numbers papers report about themselves and their baselines. They are not a leaderboard, and they do not compare across tables: each was produced by a different group, on the dataset named in its corner, with each baseline either retrained, run from released weights or quoted from another paper. Where the paper says which, it follows the method’s name (hover it for the sentence); most papers do not say. Bold is the best value in a column and underline the runner-up, our ranking rather than the paper’s own marks.

Table 2

BiATNovo: A Self-Attention based Bidirectional Peptide Sequencing Method, page 7: Results of BiATNovo and Baseline.

De novo sequencing of DIA data	Amino acid precision			Amino acid recall			Peptide precision
De novo sequencing of DIA data	OC	UTI	Plasma	OC	UTI	Plasma	OC	UTI	Plasma
DeepNovo-DIA	0.6812	0.6581	0.6379	0.6822	0.6598	0.6430	0.5240	0.5106	0.3740
BiATNovo	0.7457	0.7078	0.6628	0.7513	0.7151	0.6707	0.6147	0.5760	0.4364

Table 2

BiATNovo: An Attention-based Bidirectional De Novo Sequencing Framework for Data-Independent-Acquisition Mass Spectrometry, page 8: Comparison of DeepNovo-DIA, PepNet and BiATNovo

De novo sequencing of DIA data	Amino acid precision			Amino acid recall			Peptide precision
De novo sequencing of DIA data	OC	UTI	Plasma	OC	UTI	Plasma	OC	UTI	Plasma
DeepNovo-DIA	0.5562	0.6286	0.4354	0.5596	0.6243	0.4335	0.5098	0.5525	0.3322
BiATNovo	0.6817	0.7108	0.4525	0.6746	0.7076	0.4511	0.6439	0.6584	0.3691
PepNet			0.4008			0.4004			0.3064

Papers describing it (2)

BiATNovo: A Self-Attention based Bidirectional Peptide Sequencing Method (2023, bioRxiv, preprint)
BiATNovo: An Attention-based Bidirectional De Novo Sequencing Framework for Data-Independent-Acquisition Mass Spectrometry (2024, bioRxiv, preprint)

Authors (11)

Siyu Wu, Zhongzhi Luan, Zhenxin Fu, Qunying Wang, Tiannan Guo, Shu Yang, Binyang Li, Yuxiaomei Liu, Fangzheng Li, Jiaxing Qi, Xiaohui Liang

Seen in the charts

Back to the full map