InstaDeepAI/PXD009449

---
dataset_info:
features:
- name: sequence
dtype: large_string
- name: modified_sequence
dtype: large_string
- name: precursor_charge
dtype: int64
- name: precursor_mz
dtype: float64
- name: mz_array
large_list: float64
- name: intensity_array
large_list: float64
- name: experiment_name
dtype: large_string
- name: spectrum_id
dtype: large_string
splits:
- name: test
num_bytes: 89580538
num_examples: 41158
download_size: 59261700
dataset_size: 89580538
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
---

# Dataset Card: 21PTM dataset PXD009449 for InstaNovo-P

To assess the model performance of `InstaNovo-P` on phosphorylated peptides, we used a subset of project PXD009449 as evaluation dataset.

## Original data source:

| Field | Value |
|--------------------------|-------------------------------------------------------------------------------------------------------------------------------------------|
| Title | Systematic characterization of 21 post-translational modification using synthetic peptides |
| Description | The data presented in this study in the - context of the ProteoemTools project - is based on the synthesis of about 5000 synthetic |peptides carrying 21 different post-translational modifications to systematically characterize their chromatographic and mass spectrometric properties using multimodal LC-MS/MS.
| HostingRepository | PRIDE |
| AnnounceDate | 2024-10-22 |
| AnnouncementXML | Submission_2024-10-22_04:44:20.696.xml |
| ReviewLevel | Peer-reviewed dataset |
| DatasetOrigin | Original dataset |
| RepositorySupport | Unsupported dataset by repository |
| PrimarySubmitter | Daniel Zolg |
| SpeciesList | scientific name: Homo sapiens (Human); NCBI TaxID: 9606; |
| ModificationList | monomethylated residue; 3'-nitro-L-tyrosine; N6-malonyl-L-lysine; biotinylated residue; phosphorylated residue; acetylated residue; |dimethylated residue; iodoacetamide derivatized residue; L-citrulline; succinylated residue; ubiquitination signature dipeptidyl lysine; N6-crotonyl-L-lysine; formylated residue; monohydroxylated proline; N6,N6,N6-trimethyl-L-lysine
| Instrument | Orbitrap Fusion Lumos |
| URL | https://proteomecentral.proteomexchange.org/cgi/GetDataset?ID=PXD009449 |

## Citation:

If you use `InstaNovo-P` in your research, please cite: `InstaNovo-P: A de novo peptide sequencing model for phosphoproteomics`

```bibtex
@article {Lauridsen2025.05.14.654049,
author = {Lauridsen, Jesper and Ramasamy, Pathmanaban and Catzel, Rachel and Canbay, Vahap and Mabona, Amandla and Eloff, Kevin and Fullwood, Paul and Ferguson, Jennifer and Kirketerp-M{\o}ller, Annekatrine and Goldschmidt, Ida Sofie and Claeys, Tine and van Puyenbroeck, Sam and Lopez Carranza, Nicolas and Schoof, Erwin M. and Martens, Lennart and Van Goey, Jeroen and Francavilla, Chiara and Jenkins, Timothy Patrick and Kalogeropoulos, Konstantinos},
title = {InstaNovo-P: A de novo peptide sequencing model for phosphoproteomics},
elocation-id = {2025.05.14.654049},
year = {2025},
doi = {10.1101/2025.05.14.654049},
publisher = {Cold Spring Harbor Laboratory},
abstract = {Phosphorylation, a crucial post-translational modification (PTM), plays a central role in cellular signaling and disease mechanisms. Mass spectrometry-based phosphoproteomics is widely used for system-wide characterization of phosphorylation events. However, traditional methods struggle with accurate phosphorylated site localization, complex search spaces, and detecting sequences outside the reference database. Advances in de novo peptide sequencing offer opportunities to address these limitations, but have yet to become integrated and adapted for phosphoproteomics datasets. Here, we present InstaNovo-P, a phosphorylation specific version of our transformer-based InstaNovo model, fine-tuned on extensive phosphoproteomics datasets. InstaNovo-P significantly surpasses existing methods in phosphorylated peptide detection and phosphorylated site localization accuracy across multiple datasets, including complex experimental scenarios. Our model robustly identifies peptides with single and multiple phosphorylated sites, effectively localizing phosphorylation events on serine, threonine, and tyrosine residues. We experimentally validate our model predictions by studying FGFR2 signaling, further demonstrating that InstaNovo-P uncovers phosphorylated sites previously missed by traditional database searches. These predictions align with critical biological processes, confirming the model{\textquoteright}s capacity to yield valuable biological insights. InstaNovo-P adds value to phosphoproteomics experiments by effectively identifying biologically relevant phosphorylation events without prior information, providing a powerful analytical tool for the dissection of signaling pathways.Competing Interest StatementR.C, A.M., K.E, N.L.C., and J.V.G. are employees of InstaDeep, 5 Merchant Square, London, UK. The other authors declare no competing interests.},
URL = {https://www.biorxiv.org/content/early/2025/05/18/2025.05.14.654049},
eprint = {https://www.biorxiv.org/content/early/2025/05/18/2025.05.14.654049.full.pdf},
journal = {bioRxiv}
}
```

If you use this dataset, please cite

```bibtex
@misc{instadeep_ltd_2026,
author = { InstaDeep Ltd },
title = { PXD009449 (Revision 7676f2c) },
year = 2026,
url = { https://huggingface.co/datasets/InstaDeepAI/PXD009449 },
doi = { 10.57967/hf/7818 },
publisher = { Hugging Face }
}
```

If you use the `InstaNovo` model to generate predictions, please also cite: [InstaNovo enables diffusion-powered de novo peptide sequencing in large-scale proteomics experiments](https://doi.org/10.1038/s42256-025-01019-5)

```bibtex
@article{eloff_kalogeropoulos_2025_instanovo,
title = {InstaNovo enables diffusion-powered de novo peptide sequencing in large-scale
proteomics experiments},
author = {Eloff, Kevin and Kalogeropoulos, Konstantinos and Mabona, Amandla and Morell,
Oliver and Catzel, Rachel and Rivera-de-Torre, Esperanza and Berg Jespersen,
Jakob and Williams, Wesley and van Beljouw, Sam P. B. and Skwark, Marcin J.
and Laustsen, Andreas Hougaard and Brouns, Stan J. J. and Ljungars,
Anne and Schoof, Erwin M. and Van Goey, Jeroen and auf dem Keller, Ulrich and
Beguir, Karim and Lopez Carranza, Nicolas and Jenkins, Timothy P.},
year = 2025,
month = {Mar},
day = 31,
journal = {Nature Machine Intelligence},
doi = {10.1038/s42256-025-01019-5},
issn = {2522-5839},
url = {https://doi.org/10.1038/s42256-025-01019-5}
}
```

数据集信息：
特征：
- 名称：sequence（序列），数据类型：large_string（大字符串）
- 名称：modified_sequence（修饰序列），数据类型：large_string（大字符串）
- 名称：precursor_charge（前体电荷），数据类型：int64（64位整数）
- 名称：precursor_mz（前体质荷比），数据类型：float64（64位浮点数）
- 名称：mz_array（质荷比数组），数据类型：large_list（大列表），元素类型：float64（64位浮点数）
- 名称：intensity_array（强度数组），数据类型：large_list（大列表），元素类型：float64（64位浮点数）
- 名称：experiment_name（实验名称），数据类型：large_string（大字符串）
- 名称：spectrum_id（谱图ID），数据类型：large_string（大字符串）
数据集划分：
- 名称：test（测试集），字节占用：89580538，样本数量：41158
下载大小：59261700，数据集总大小：89580538
配置项：
- 配置名称：default（默认配置），数据文件：
- 划分：test（测试集），路径：data/test-*

# 数据集卡片：用于InstaNovo-P的21种翻译后修饰（post-translational modification, PTM）数据集PXD009449

为评估`InstaNovo-P`在磷酸化肽段上的模型性能，我们选取了项目PXD009449的一个子集作为评估数据集。

## 原始数据源：

| 字段名 | 取值 |
|--------------------------|-------------------------------------------------------------------------------------------------------------------------------------------|
| 标题 | 利用合成肽段系统表征21种翻译后修饰 |
| 描述 | 本研究呈现的数据隶属于ProteomeTools（蛋白质组工具）项目，基于约5000条携带21种不同翻译后修饰的合成肽段的合成实验，旨在通过多模态液相色谱-串联质谱（LC-MS/MS）系统表征其色谱及质谱特性。 |
| 托管仓库 | PRIDE |
| 公告日期 | 2024-10-22 |
| 公告XML文件 | Submission_2024-10-22_04:44:20.696.xml |
| 评审级别 | 同行评审数据集 |
| 数据集来源 | 原始数据集 |
| 仓库支持情况 | 仓库不支持该数据集 |
| 主要提交者 | Daniel Zolg |
| 物种列表 | 科学名称：智人（Homo sapiens，人类）；NCBI分类ID：9606 |
| 修饰列表 | 单甲基化残基；3'-硝基-L-酪氨酸；N6-丙二酰-L-赖氨酸；生物素化残基；磷酸化残基；乙酰化残基；二甲基化残基；碘乙酰胺衍生化残基；L-瓜氨酸；琥珀酰化残基；泛素化特征二肽基赖氨酸；N6-巴豆酰-L-赖氨酸；甲酰化残基；单羟基化脯氨酸；N6,N6,N6-三甲基-L-赖氨酸 |
| 仪器 | Orbitrap Fusion Lumos（轨道阱融合卢莫斯质谱仪） |
| 网址 | https://proteomecentral.proteomexchange.org/cgi/GetDataset?ID=PXD009449 |

## 引用：

若在研究中使用`InstaNovo-P`，请引用以下文献：《InstaNovo-P: 用于磷酸化蛋白质组学的从头肽段测序模型》

bibtex
@article {Lauridsen2025.05.14.654049,
作者 = {Lauridsen, Jesper 及合作者},
标题 = {InstaNovo-P：用于磷酸化蛋白质组学的从头肽段测序模型},
电子定位ID = {2025.05.14.654049},
年份 = {2025},
doi = {10.1101/2025.05.14.654049},
出版者 = {冷泉港实验室},
摘要 = {磷酸化作为一种关键的翻译后修饰（PTM），在细胞信号传导与疾病机制中发挥核心作用。基于质谱的磷酸化蛋白质组学被广泛用于磷酸化事件的系统级表征。然而，传统方法在精准磷酸化位点定位、复杂搜索空间以及检测参考数据库外的序列等方面存在局限。从头肽段测序技术的进展为解决这些局限提供了契机，但尚未被整合并适配至磷酸化蛋白质组学数据集。本文中我们提出InstaNovo-P，这是我们基于Transformer的InstaNovo模型的磷酸化特异性版本，在大规模磷酸化蛋白质组学数据集上进行了微调。InstaNovo-P在多个数据集（包括复杂实验场景）上的磷酸化肽段检测与磷酸化位点定位精度上显著优于现有方法。我们的模型能够稳健地识别带有单个和多个磷酸化位点的肽段，有效定位丝氨酸、苏氨酸和酪氨酸残基上的磷酸化事件。我们通过研究FGFR2信号通路对模型预测进行了实验验证，进一步证明InstaNovo-P能够发现传统数据库搜索遗漏的磷酸化位点。这些预测与关键生物学过程相符，证实了该模型能够产生有价值的生物学见解。InstaNovo-P无需先验信息即可有效识别具有生物学意义的磷酸化事件，为磷酸化蛋白质组学实验提供了价值，为解析信号通路提供了强大的分析工具。
利益冲突声明：R.C、A.M、K.E、N.L.C和J.V.G为英国伦敦InstaDeep公司员工，其余作者声明无利益冲突。},
网址 = {https://www.biorxiv.org/content/early/2025/05/18/2025.05.14.654049},
预印本链接 = {https://www.biorxiv.org/content/early/2025/05/18/2025.05.14.654049.full.pdf},
期刊 = {bioRxiv}
}

若使用本数据集，请引用：
bibtex
@misc{instadeep_ltd_2026,
作者 = { InstaDeep有限公司 },
标题 = { PXD009449（修订版7676f2c） },
年份 = 2026,
url = { https://huggingface.co/datasets/InstaDeepAI/PXD009449 },
doi = { 10.57967/hf/7818 },
出版者 = { Hugging Face }
}

若使用`InstaNovo`模型生成预测结果，请同时引用：[InstaNovo enables diffusion-powered de novo peptide sequencing in large-scale proteomics experiments](https://doi.org/10.1038/s42256-025-01019-5)

bibtex
@article{eloff_kalogeropoulos_2025_instanovo,
标题 = {InstaNovo助力大规模蛋白质组学实验中基于扩散模型的从头肽段测序},
作者 = {Eloff, Kevin 及合作者},
年份 = 2025,
月份 = {3月},
日期 = 31,
期刊 = {Nature Machine Intelligence},
doi = {10.1038/s42256-025-01019-5},
issn = {2522-5839},
url = {https://doi.org/10.1038/s42256-025-01019-5}
}