2021 年,人工智能的应用,在制药、医疗、合成生物学等领域取得了丰硕进展。

以下为ScienceAI对相关领域报道的年度回顾,主要分为两部分,第一部分是对于研究的报道,第二部分是领域内各类专家的言论访谈、相关企业的报道。

研究

可对药物分子进行表征的几何深度学习

Nature Machine Intelligence | Geometric deep learning on molecular representations

论文链接:https://www.nature.com/articles/s42256-021-00418-8

人工智能助力疾病研究与医疗诊断

Nature | Disease variant prediction with deep generative models of evolutionary data

论文链接:https://www.nature.com/articles/s41586-021-04043-8

Nature Communications | A machine and human reader study on AI diagnosis model safety under attacks of adversarial images

论文链接:https://www.nature.com/articles/s41467-021-27577-x

arXiv 预印平台 | Multimodal Representation Learning via Maximization of Local Mutual Information

论文链接:https://arxiv.org/abs/2103.04537

PLOS GENETICS | Machine learning to predict the source of campylobacteriosis using whole genome data

论文链接:https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1009436

五项研究,人工智能助力疾病发现以及药物开发

Nature Genetics | Deep learning enables genetic analysis of the human thoracic aorta

论文链接:https://www.nature.com/articles/s41588-021-00962-4

Bioinformatics | Cancer subtype identification by consensus guided graph autoencoders

论文链接:https://academic.oup.com/bioinformatics/article-abstract/37/24/4779/6325018?redirectedFrom=fulltext

Bioinformatics | BioERP: biomedical heterogeneous network-based self-supervised representation learning approach for entity relationship predictions

论文链接:https://academic.oup.com/bioinformatics/article-abstract/37/24/4793/6332000?redirectedFrom=fulltext

Journal of Pharmacokinetics and Pharmacodynamics | Machine learning-guided, big data-enabled, biomarker-based systems pharmacology: modeling the stochasticity of natural history and disease progression

论文链接:https://link.springer.com/article/10.1007/s10928-021-09786-5

Bioinformatics | DTI-Voodoo: machine learning over interaction networks and ontology-based background knowledge predicts drug–target interactions

论文链接:https://academic.oup.com/bioinformatics/article/37/24/4835/6329632

深度学习模型可快速预测「类药物分子」的 3D 形状

arXiv 预印平台 | GeoMol: Torsional Geometric Generation of Molecular 3D Conformer Ensembles

论文链接:https://arxiv.org/abs/2106.07802

人工智能发现人类蛋白质具有杀灭超级细菌的潜力

Nature Biomedical Engineering | Mining for encrypted peptide antibiotics in the human proteome

论文链接:https://www.nature.com/articles/s41551-021-00801-1

近期五项研究,人工智能助力疾病治疗

Front. Oncol. | Novel Non-Invasive Radiomic Signature on CT Scans Predicts Response to Platinum-Based Chemotherapy and Is Prognostic of Overall Survival in Small Cell Lung Cancer

论文链接:https://doi.org/10.3389/fonc.2021.744724

Scientific Reports | The impact of recency and adequacy of historical information on sepsis predictions using machine learning

论文链接:https://www.nature.com/articles/s41598-021-00220-x

Melanoma Research | The potential of using artificial intelligence to improve skin cancer diagnoses in Hawai‘i’s multiethnic population

论文链接:https://journals.lww.com/melanomaresearch/Abstract/2021/12000/The_potential_of_using_artificial_intelligence_to.2.aspx

PNAS | Transformational machine learning: Learning how to learn from many related scientific problems

论文链接:https://www.pnas.org/content/118/49/e2108013118

Nature Biomedical Engineering | Quantitative imaging of apoptosis following oncolytic virotherapy by magnetic resonance fingerprinting aided by deep learning

论文链接:https://www.nature.com/articles/s41551-021-00809-7

新型人工基因组DNA可以在细胞外复制和进化

ACS Synthetic Biology | Continuous Cell-Free Replication and Evolution of Artificial Genomic DNA in a Compartmentalized Gene Expression System

论文链接:https://pubs.acs.org/doi/10.1021/acssynbio.1c00430

近期六项人工智能研究助力癌症治疗

人工智能助力缉毒,深度生成模型可自动解析新型精神活性物质的结构

Nature Machine Intelligence | A deep generative model enables automated structure elucidation of novel psychoactive substances

论文链接:https://www.nature.com/articles/s42256-021-00407-x

近期四项研究:AI助力癌症检测与抗癌药物研发,从实验室到企业

Journal of Oncology Research | ELAFT: An Ensemble-based Machine-learning Algorithm that Predicts Anti-cancer Drug Responses with High Accuracy

论文链接:https://escires.com/articles/JOR-4-111.pdf

Nature Communications | Accurate recognition of colorectal cancer with semi-supervised deep learning on pathological images

论文链接:https://www.nature.com/articles/s41467-021-26643-8

BioDynaMo(已开源):一个研究生物过程的新计算平台

Bioinformatics | BioDynaMo: a modular platform for high-performance agent-based simulation

论文链接:https://academic.oup.com/bioinformatics/advance-article/doi/10.1093/bioinformatics/btab649/6371176

机器学习通过分析水平基因转移,预测抗生素耐药性传播

Science Advances | Functions predict horizontal gene transfer and the emergence of antibiotic resistance

论文链接:https://www.science.org/doi/10.1126/sciadv.abj5056

AI助力癌症治疗与预防,近期领域内的五项新成果

Nature | Biologically informed deep neural network for prostate cancer discovery

论文链接:https://www.nature.com/articles/s41586-021-03922-4

Nature Machine Intelligence | Deep learning-based prediction of the T cell receptor–antigen binding specificity

论文链接:https://www.nature.com/articles/s42256-021-00383-2

Nature Communications | Robust whole slide image analysis for cervical cancer screening using deep learning

论文链接:https://www.nature.com/articles/s41467-021-25296-x

Nature Communications | Artificial intelligence system reduces false-positive findings in the interpretation of breast ultrasound exams

论文链接:https://www.nature.com/articles/s41467-021-26023-2

Radiology | Deep Learning for Automated Triaging of 4581 Breast MRI Examinations from the DENSE Trial

论文链接:https://pubs.rsna.org/doi/10.1148/radiol.2021203960

机器学习揭示了农业和医学中的「重要基因」

Nature Communications | Evolutionarily informed machine learning enhances the power of predictive gene-to-phenotype relationships

论文链接:https://www.nature.com/articles/s41467-021-25893-w

将人工智能和on-chip合成相结合,助力新药设计

Scinece Advances | Combining generative artificial intelligence and on-chip synthesis for de novo drug design

论文链接:https://www.science.org/doi/10.1126/sciadv.abg3338

机器学习发现新序列,促进药物输送,「杜氏肌营养不良」治疗再现曙光?

Nature Chemistry | Deep learning to design nuclear-targeting abiotic miniproteins

论文链接:https://www.nature.com/articles/s41557-021-00766-3

癌细胞基因以及纳米治疗方法,机器学习助力癌症研究

Nature | In silico saturation mutagenesis of cancer genes

论文链接:https://www.nature.com/articles/s41586-021-03771-1

魔法还是巫术,迈向合成细胞周期

Nature Communications | Towards a synthetic cell cycle

论文链接:https://www.nature.com/articles/s41467-021-24772-8

张锋团队新进展|利用人类蛋白质向细胞靶向提供分子药物

Science | Mammalian retrovirus-like protein PEG10 packages its own mRNA and can be pseudotyped for mRNA delivery

论文链接:https://science.sciencemag.org/content/373/6557/882

综述:药物发现中的机器学习

Artificial Intelligence Review | Machine Learning in Drug Discovery: A Review

论文链接:https://link.springer.com/article/10.1007%2Fs10462-021-10058-4

X光片还能显示分子构成?杜克大学研发混合X射线扫描仪加速癌症诊断

Scientific Reports | X-ray fan beam coded aperture transmission and diffraction imaging for fast material analysis

论文链接:https://doi.org/10.1038/s41598-021-90163-0

隐藏于药物研发过程中的合成生物学

SLAS DISCOVERY | A Perspective on Synthetic Biology in Drug Discovery and Development—Current Impact and Future Opportunities

论文链接:https://journals.sagepub.com/doi/abs/10.1177/24725552211000669?journalCode=jbxb

RNN 用于生物医学全息成像,速度加快50倍

ACS Photonics | Holographic Image Reconstruction with Phase Recovery and Autofocusing Using Recurrent Neural Networks 

论文链接:https://pubs.acs.org/doi/10.1021/acsphotonics.1c00337

AI设计出更好的抗体药物,可降低临床风险

Nature Biomedical Engineering | Optimization of therapeutic antibodies by predicting antigen specificity from antibody sequence via deep learning

论文链接:https://www.nature.com/articles/s41551-021-00699-9

超过突变总和:AI识别出165个新癌症基因

Nature Machine Intelligence  | Integration of multiomics data with graph convolutional networks to identify new cancer genes and their associated molecular mechanisms

论文链接:http://dx.doi.org/10.1038/s42256-021-00325-y

MIT化学家开发AI应用,为药物发现提速

The Journal of Physical Chemistry Letters | DeepBAR: A Fast and Exact Method for Binding Free Energy Computation

论文链接:https://doi.org/10.1021/acs.jpclett.1c00189

强化学习模拟自适应免疫系统,或能带来新的免疫学见解

Physical Review Research | Understanding Adaptive Immune System as Reinforcement Learning

综述论文:人工智能在药物研发三大环节中的应用

论文链接:https://www.deepdyve.com/lp/sage/artificial-intelligence-effecting-a-paradigm-shift-in-drug-development-pfYlpklDRo

内容中包含的图片若涉及版权问题,请及时与我们联系删除