Anqi Liu

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

CLCLSA: Cross-omics Linked embedding with Contrastive Learning and Self Attention for multi-omics integration with incomplete multi-omics data

Research Square (Research Square) (2023)

Weihua Zhou Zhao Chen Anqi Liu Xiao Zhang Xuewei Cao

Integration of heterogeneous and high-dimensional multi-omics data is becoming increasingly important in understanding genetic data. Each omics technique only provides a limited view of the underlying biological process and integrating heterogeneous omics layers simultaneously would lead to a more comprehensive and detailed understanding of diseases and phenotypes. However, one obstacle faced when performing multi-omics data integration is the existence of unpaired multi-omics data due to instrument sensitivity and cost. Studies may fail if certain aspects of the subjects are missing or incomplete. In this paper, we propose a deep learning method for multi-omics integration with incomplete data by Cross-omics Linked unified embedding with Contrastive Learning and Self Attention (CLCLSA). Utilizing complete multi-omics data as supervision, the model employs cross-omics autoencoders to learn the feature representation across different types of biological data. The multi-omics contrastive learning, which is used to maximize the mutual information between different types of omics, is employed before latent feature concatenation. In addition, the feature-level self-attention and omics-level self-attention are employed to dynamically identify the most informative features for multi-omics data integration. Extensive experiments were conducted on four public multi-omics datasets. The experimental results indicated that the proposed CLCLSA outperformed the state-of-the-art approaches for multi-omics data classification using incomplete multi-omics data.

Omics

Feature (linguistics)

10.21203/rs.3.rs-2768563/v1

Cite

Citations (1)

A New Type of Pump's Design and Research

Oil Field Equipment (2007)

Anqi Liu

To some problem,such as a short service life,being difficult to change and repair,slurry leaking in sealant of axis,we designed a new pump which interspaces is filled with air between sealant and slurry.Then we solved the problem radically.And we analyzed the velocity and pressure of liquid by CFX,to prove the feasibility of the new configuration.This new pump could conquer a series of problems which arose by slurry leaking.For instance, avoided slurry leaking,eliminated the pollution,economized the cost of drilling,improved the efficiency of pump and increased the mechanical rev.

Oil pump

Service life

Lost Circulation

Source

Cite

Citations (0)

CLCLSA: Cross-omics linked embedding with contrastive learning and self attention for integration with incomplete multi-omics data

Computers in Biology and Medicine (2024)

Zhao Chen Anqi Liu X Zhang Xuewei Cao Zhengming Ding

Omics

10.1016/j.compbiomed.2024.108058

Cite

Citations (4)

Utilizing carbon dioxide from refinery for methanol and electricity co-production: System design and economic assessment

E3S Web of Conferences (2023)

Qian Ma Yuxi Wang Xu Zhang Xiaodan Yu Anqi Liu

Mitigating greenhouse gas (GHG) emissions to curb climate change has become a consensus among international community. Refining is one of the major industries with high energy consumption and high emissions, which is responsible for 4-10% of global carbon dioxide (CO 2 ) emissions and approximately 25% is generated by fluid catalytic cracking (FCC) units. The flue gas discharged from FCC units has a high CO 2 content, presenting potential for methanol and electricity co-production production when the methane in the dry gas is considered. To unlock this green chance for enterprises, we designed a methanol production system that uses the CO 2 recovered from the flue gas of fluid catalytic cracking (FCC) unit in a refinery and the CH4 separated from dry gas as feed stock (hydrogen to be added from an internal hydrogen production unit of the refinery). We analyzed economic feasibility of the process, the results show that the developed system is economically feasible for annual methanol yields of 1.0-2.5 Mt and the internal rate of return increases by 8.3%.

Refinery

Fluid catalytic cracking

10.1051/e3sconf/202337502012

Cite

Citations (0)

Novel Process for Carbon Cycle Utilization from Industrial Flue Gas into Methanol

E3S Web of Conferences (2023)

Qian Ma Xu Zhang Xiaodan Yu Anqi Liu Bo Yuan

Reduction of greenhouse gas (GHG) emissions for mitigating climate changes is globally necessary. As a major GHG emitter, refinery sector is responsible for 14.1% of China’s CO 2 emission in 2019. The flue gas associated with petroleum refining has high CO 2 content, presenting potentials for methanol and electricity co-production when the methane and hydrogen in dry gas are considered. To unlock the green opportunity for refineries, in this study we employed Aspen Plus to develop the process of methanol synthesis by recovering CO 2 in the flue gas of fluid catalytic cracking (FCC) unit and the methane and hydrogen in dry gas. The results show that the new co-production process increases the energy efficiency of the FCC unit by 2.4%. Compared with traditionally natural gas-based methanol production, the developed process enables annual CO 2 mitigation of 1.3 million tons.

Refinery

10.1051/e3sconf/202340603008

Cite

Citations (0)

Analysis and prediction of distribution network load characteristics based on data mining

Journal of Physics Conference Series (2024)

Peng Ye Yufei Yuan Na Zheng Jun Liu Shiyu Li

Abstract Load forecasting helps the power department make annual, scheduling, and maintenance plans. In this paper, we take the distribution network of Shenyang as the research object, use the spectral clustering algorithm to cluster the load data, and analyze the correlation between the load and external factors through the correlation and gray correlation method. Moreover, we use the BO - TPE - XGBoost model for load forecasting and compare it with other methods. The results show that this method significantly improves the forecasting accuracy.

10.1088/1742-6596/2896/1/012040

Cite

Citations (0)

Utilizing CO₂ from refinery for methanol and electricity coproduction: System assessment

E3S Web of Conferences (2023)

Qian Ma Xiaodan Yu Song Xia Xu Zhang Anqi Liu

As one of the major energy-consuming and high-emission industries, the refineries account for 10% of global industrial carbon emissions, of which about 25% is emitted by fluid catalytic cracking (FCC) process. Therefore, CO 2 capture from FCC flue gas create the possibility for low-carbon refineries. We developed a novel CO 2 utilization process to methanol production modeling using Aspen Plus. Meanwhile, organic Rankine Cycle (ORC) power generation technology was coupled to recover the waste heat of the system. A five-million-ton FCC unit in China was selected as a case study. We conducted some analysis for the process, the results show that the developed system boosts the energy efficiency of the FCC unit by 2.8%. The annual capacity of the waste heat power generation unit is 4.8 GWh, with a thermal efficiency of 5.9%.

Refinery

Organic Rankine Cycle

Fluid catalytic cracking

10.1051/e3sconf/202337502011

Cite

Citations (0)