Embeddings from protein language models predict conservation and variant effects.
Embeddings from protein language models predict conservation and variant effects.
复制标题
DOI：
10.1007/s00439-021-02411-y
复制
发表时间：
2022-10
期刊：
影响因子：
5.3
通讯作者：
Rost B
中科院分区：
文献类型：
--
作者：
Marquet C;Heinzinger M;Olenyi T;Dallago C;Erckert K;Bernhofer M;Nechaev D;Rost B
SARS-CoV-2变异体的出现强调了对允许解释单个氨基酸变异体（SAV）对蛋白质功能的影响的工具的需求。虽然深度突变扫描（DMS）集继续扩大我们对单个蛋白质突变景观的理解，但结果继续挑战分析。蛋白质语言模型（pLM）使用最新的深度学习（DL）算法来利用不断增长的蛋白质序列数据库。这些方法学习从整个序列区域的背景中预测缺失或掩蔽的氨基酸。在这里，我们使用pLM表示（嵌入）来预测序列保守性和SAV效应，而无需多重序列比对（MSA）。单独的嵌入与使用MSA的ConSeq几乎一样准确地从单个序列预测残基保守性（ProtT 5嵌入的双态马修斯相关系数-MCC-为0.596 ± 0.006对ConSeq的0.608 ± 0.006）。将保守性预测沿着BL 0 SUM 62取代评分和pLM掩模重建概率输入到用于变体效应预测评分（VESPA）的简化逻辑回归（LR）系综中，预测SAV效应幅度，而无需对DMS数据进行任何优化。将39个DMS实验的标准集的预测与其他方法（包括ESM-1v、DeepSequence和GEMME）表明，我们的方法与使用MSA输入的最先进（SOTA）方法具有竞争力。没有方法优于所有其他方法，既不一致，也不具有统计学显著性，独立于所应用的性能指标（斯皮尔曼和皮尔逊相关性）。最后，我们研究了四种人类蛋白质的DMS实验的二元效应预测。总的来说，基于嵌入的方法已经变得与依赖于MSA的方法竞争，以计算/能量成本的一小部分用于SAV效应预测。我们的方法在40分钟内预测了整个人类蛋白质组（约20 k蛋白质）在Nvidia Quadro RTX 8000上的SAV效应。所有方法和数据集都可以通过bioembeddings.com、https：//github.com/Rostlab/VESPA和PredictProtein免费获得，供本地和在线执行。在线版本包含补充材料，可通过10.1007/s 00439 -021-02411-y获得。
The emergence of SARS-CoV-2 variants stressed the demand for tools allowing to interpret the effect of single amino acid variants (SAVs) on protein function. While Deep Mutational Scanning (DMS) sets continue to expand our understanding of the mutational landscape of single proteins, the results continue to challenge analyses. Protein Language Models (pLMs) use the latest deep learning (DL) algorithms to leverage growing databases of protein sequences. These methods learn to predict missing or masked amino acids from the context of entire sequence regions. Here, we used pLM representations (embeddings) to predict sequence conservation and SAV effects without multiple sequence alignments (MSAs). Embeddings alone predicted residue conservation almost as accurately from single sequences as ConSeq using MSAs (two-state Matthews Correlation Coefficient—MCC—for ProtT5 embeddings of 0.596 ± 0.006 vs. 0.608 ± 0.006 for ConSeq). Inputting the conservation prediction along with BLOSUM62 substitution scores and pLM mask reconstruction probabilities into a simplistic logistic regression (LR) ensemble for Variant Effect Score Prediction without Alignments (VESPA) predicted SAV effect magnitude without any optimization on DMS data. Comparing predictions for a standard set of 39 DMS experiments to other methods (incl. ESM-1v, DeepSequence, and GEMME) revealed our approach as competitive with the state-of-the-art (SOTA) methods using MSA input. No method outperformed all others, neither consistently nor statistically significantly, independently of the performance measure applied (Spearman and Pearson correlation). Finally, we investigated binary effect predictions on DMS experiments for four human proteins. Overall, embedding-based methods have become competitive with methods relying on MSAs for SAV effect prediction at a fraction of the costs in computing/energy. Our method predicted SAV effects for the entire human proteome (~ 20 k proteins) within 40 min on one Nvidia Quadro RTX 8000. All methods and data sets are freely available for local and online execution through bioembeddings.com, https://github.com/Rostlab/VESPA, and PredictProtein. The online version contains supplementary material available at 10.1007/s00439-021-02411-y.
登录
查看更多内容
影响因子：
64.8
作者：
Findlay GM;Daza RM;Martin B;Zhang MD;Leith AP;Gasperini M;Janizek JD;Huang X;Starita LM;Shendure J
通讯作者：
Shendure J
影响因子：
48
作者：
Fowler, Douglas M.;Fields, Stanley
通讯作者：
Fields, Stanley
DOI：
10.1093/bioinformatics/bts565
发表时间：
2012-12-01
期刊：
Bioinformatics (Oxford, England)
影响因子：
--
作者：
Fu L;Niu B;Zhu Z;Wu S;Li W
通讯作者：
Li W
影响因子：
14.9
作者：
Bromberg Y;Rost B
通讯作者：
Rost B
影响因子：
48
作者：
Alley, Ethan C.;Khimulya, Grigory;Church, George M.
通讯作者：
Church, George M.