Lei Fang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Lei Fang has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
Jian-Guang Lou
2 shared publications- Jian-Guang Lou
Microsoft Research Asia, Beijing, China
2 shared publications - Junren Li
BNLMS, College of Chemistry and Molecular Engineering, Peking University, Beijing, China
2 shared publications - Albert Neutzner
Department of Biomedicine, University of Basel, Basel, Switzerland
1 shared publication - Bei Chen
Center of Visual Information and Applications Institute of Artificial Intelligence and Robotics National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Xi'an Jiaotong University ♣ Microsoft Corporation ♡ Northeastern University
1 shared publication - Biao Liu
Beijing Jiaotong University
1 shared publication
Latest Publications
- Open Access
- 2 February 2024
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 3 (3) , 482-490
Retro-BLEU: quantifying chemical plausibility of retrosynthesis routes through reaction template sequence analysis
Retro-BLEU is a statistical metric to evaluate the plausibility of model-generated retrosynthesis routes based on reaction template sequences analysis.
- 2 February 2024
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 3 (3) , 482-490
Open AccessRetro-BLEU: quantifying chemical plausibility of retrosynthesis routes through reaction template sequence analysis
Retro-BLEU is a statistical metric to evaluate the plausibility of model-generated retrosynthesis routes based on reaction template sequences analysis.
- Open Access
Retro-BLEU: Quantifying Chemical Plausibility of Retrosynthesis Routes through Reaction Template Sequence Analysis
Computer-assisted methods have emerged as valuable tools for retrosynthesis analysis. However, quantifying the plausibility of generated retrosynthesis routes remains a challenging task. We introduce Retro-BLEU, a statistical metric adapted from the well-established BLEU score in machine translation, to evaluate the plausibility of retrosynthesis routes based on reaction template sequences analysis. We demonstrate the effectiveness of Retro-BLEU by applying it to a diverse set of retrosynthesis routes generated by state-of-the-art algorithms and compare the performance with other evaluation metrics. The results show that Retro-BLEU is capable of differentiating between plausible and implausible routes. Furthermore, we provide insights into the strengths and weaknesses of Retro-BLEU, paving the way for future developments and improvements in this field.
Open AccessRetro-BLEU: Quantifying Chemical Plausibility of Retrosynthesis Routes through Reaction Template Sequence Analysis
Computer-assisted methods have emerged as valuable tools for retrosynthesis analysis. However, quantifying the plausibility of generated retrosynthesis routes remains a challenging task. We introduce Retro-BLEU, a statistical metric adapted from the well-established BLEU score in machine translation, to evaluate the plausibility of retrosynthesis routes based on reaction template sequences analysis. We demonstrate the effectiveness of Retro-BLEU by applying it to a diverse set of retrosynthesis routes generated by state-of-the-art algorithms and compare the performance with other evaluation metrics. The results show that Retro-BLEU is capable of differentiating between plausible and implausible routes. Furthermore, we provide insights into the strengths and weaknesses of Retro-BLEU, paving the way for future developments and improvements in this field.
- Open Access
- 8 June 2023
- Published by Springer Nature in Journal of Cheminformatics
- Vol. 15 (1) , 1-13
RetroRanker: leveraging reaction changes to improve retrosynthesis prediction through re-ranking
Retrosynthesis is an important task in organic chemistry. Recently, numerous data-driven approaches have achieved promising results in this task. However, in practice, these data-driven methods might lead to sub-optimal outcomes by making predictions based on the training data distribution, a phenomenon we refer as frequency bias. For example, in template-based approaches, low-ranked predictions are typically generated by less common templates with low confidence scores which might be too low to be comparable, and it is observed that recorded reactants can be among these low-ranked predictions. In this work, we introduce RetroRanker, a ranking model built upon graph neural networks, designed to mitigate the frequency bias in predictions of existing retrosynthesis models through re-ranking. RetroRanker incorporates potential reaction changes of each set of predicted reactants in obtaining the given product to lower the rank of chemically unreasonable predictions. The predicted re-ranked results on publicly available retrosynthesis benchmarks demonstrate that we can achieve improvement on most state-of-the-art models with RetroRanker. Our preliminary studies also indicate that RetroRanker can enhance the performance of multi-step retrosynthesis.
- 8 June 2023
- Published by Springer Nature in Journal of Cheminformatics
- Vol. 15 (1) , 1-13
Open AccessRetroRanker: leveraging reaction changes to improve retrosynthesis prediction through re-ranking
Retrosynthesis is an important task in organic chemistry. Recently, numerous data-driven approaches have achieved promising results in this task. However, in practice, these data-driven methods might lead to sub-optimal outcomes by making predictions based on the training data distribution, a phenomenon we refer as frequency bias. For example, in template-based approaches, low-ranked predictions are typically generated by less common templates with low confidence scores which might be too low to be comparable, and it is observed that recorded reactants can be among these low-ranked predictions. In this work, we introduce RetroRanker, a ranking model built upon graph neural networks, designed to mitigate the frequency bias in predictions of existing retrosynthesis models through re-ranking. RetroRanker incorporates potential reaction changes of each set of predicted reactants in obtaining the given product to lower the rank of chemically unreasonable predictions. The predicted re-ranked results on publicly available retrosynthesis benchmarks demonstrate that we can achieve improvement on most state-of-the-art models with RetroRanker. Our preliminary studies also indicate that RetroRanker can enhance the performance of multi-step retrosynthesis.
- Open Access
- 28 April 2023
- Published by Springer Nature in Nature Communications
- Vol. 14 (1) , 1-14
Single-step retrosynthesis prediction by leveraging commonly preserved substructures
Retrosynthesis analysis is an important task in organic chemistry with numerous industrial applications. Previously, machine learning approaches employing natural language processing techniques achieved promising results in this task by first representing reactant molecules as strings and subsequently predicting reactant molecules using text generation or machine translation models. Chemists cannot readily derive useful insights from traditional approaches that rely largely on atom-level decoding in the string representations, because human experts tend to interpret reactions by analyzing substructures that comprise a molecule. It is well-established that some substructures are stable and remain unchanged in reactions. In this paper, we developed a substructure-level decoding model, where commonly preserved portions of product molecules were automatically extracted with a fully data-driven approach. Our model achieves improvement over previously reported models, and we demonstrate that its performance can be boosted further by enhancing the accuracy of these substructures. Analyzing substructures extracted from our machine learning model can provide human experts with additional insights to assist decision-making in retrosynthesis analysis.
- 28 April 2023
- Published by Springer Nature in Nature Communications
- Vol. 14 (1) , 1-14
Open AccessSingle-step retrosynthesis prediction by leveraging commonly preserved substructures
Retrosynthesis analysis is an important task in organic chemistry with numerous industrial applications. Previously, machine learning approaches employing natural language processing techniques achieved promising results in this task by first representing reactant molecules as strings and subsequently predicting reactant molecules using text generation or machine translation models. Chemists cannot readily derive useful insights from traditional approaches that rely largely on atom-level decoding in the string representations, because human experts tend to interpret reactions by analyzing substructures that comprise a molecule. It is well-established that some substructures are stable and remain unchanged in reactions. In this paper, we developed a substructure-level decoding model, where commonly preserved portions of product molecules were automatically extracted with a fully data-driven approach. Our model achieves improvement over previously reported models, and we demonstrate that its performance can be boosted further by enhancing the accuracy of these substructures. Analyzing substructures extracted from our machine learning model can provide human experts with additional insights to assist decision-making in retrosynthesis analysis.
- Open Access
- 30 August 2021
- Published by International Speech Communication Association
- p. 1499-1503
A Weight Moving Average Based Alternate Decoupled Learning Algorithm for Long-Tailed Language Identification
- 30 August 2021
- Published by International Speech Communication Association
- p. 1499-1503
Open AccessA Weight Moving Average Based Alternate Decoupled Learning Algorithm for Long-Tailed Language Identification
- 1 August 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 1, 64-69
Traditional Chinese Medicine knowledge Service based on Semi-Supervised BERT-BiLSTM-CRF Model
Most of Traditional Chinese Medicine (TCM) data and ancient records exist in the form of books. The unstructured medical information is the foundation for building TCM knowledge service. The existing methods are not accurate enough to solve TCM named entity recognition and require a lot of manual labeling data. This paper proposes a semi-supervised embedded Semi-BERT-BiLSTM-CRF model. Based on the book “Diagnosis of Traditional Chinese Medicine in Traditional Chinese Medicine”, we select the physical features from the cleaned-up text information according to the characteristics of Chinese medicine classics, and then use a small amount of labeled data to train the BERT-BiLSTM-CRF model. The obtained model is used to predict unlabeled data and obtain pseudo-label data. The pseudo-label and labeled data are used as a training set for model training. Experiments show that TCM entity recognition accuracy of this method reaches 81.24%, which effectively improves the TCM entity recognition accuracy and reduces the manual labeling work. The results of this research can be applied to scenarios such as auxiliary diagnosis of TCM and expert system after subsequent improvement and transformation.
- 1 August 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 1, 64-69
Traditional Chinese Medicine knowledge Service based on Semi-Supervised BERT-BiLSTM-CRF Model
Most of Traditional Chinese Medicine (TCM) data and ancient records exist in the form of books. The unstructured medical information is the foundation for building TCM knowledge service. The existing methods are not accurate enough to solve TCM named entity recognition and require a lot of manual labeling data. This paper proposes a semi-supervised embedded Semi-BERT-BiLSTM-CRF model. Based on the book “Diagnosis of Traditional Chinese Medicine in Traditional Chinese Medicine”, we select the physical features from the cleaned-up text information according to the characteristics of Chinese medicine classics, and then use a small amount of labeled data to train the BERT-BiLSTM-CRF model. The obtained model is used to predict unlabeled data and obtain pseudo-label data. The pseudo-label and labeled data are used as a training set for model training. Experiments show that TCM entity recognition accuracy of this method reaches 81.24%, which effectively improves the TCM entity recognition accuracy and reduces the manual labeling work. The results of this research can be applied to scenarios such as auxiliary diagnosis of TCM and expert system after subsequent improvement and transformation.
- 26 August 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Visualization and Computer Graphics
- Vol. 26 (1) , 906-916
Text-to-Viz: Automatic Generation of Infographics from Proportion-Related Natural Language Statements
Combining data content with visual embellishments, infographics can effectively deliver messages in an engaging and memorable manner. Various authoring tools have been proposed to facilitate the creation of infographics. However, creating a professional infographic with these authoring tools is still not an easy task, requiring much time and design expertise. Therefore, these tools are generally not attractive to casual users, who are either unwilling to take time to learn the tools or lacking in proper design expertise to create a professional infographic. In this paper, we explore an alternative approach: to automatically generate infographics from natural language statements. We first conducted a preliminary study to explore the design space of infographics. Based on the preliminary study, we built a proof-of-concept system that automatically converts statements about simple proportionrelated statistics to a set of infographics with pre-designed styles. Finally, we demonstrated the usability and usefulness of the system through sample results, exhibits, and expert reviews.
- 26 August 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Visualization and Computer Graphics
- Vol. 26 (1) , 906-916
Text-to-Viz: Automatic Generation of Infographics from Proportion-Related Natural Language Statements
Combining data content with visual embellishments, infographics can effectively deliver messages in an engaging and memorable manner. Various authoring tools have been proposed to facilitate the creation of infographics. However, creating a professional infographic with these authoring tools is still not an easy task, requiring much time and design expertise. Therefore, these tools are generally not attractive to casual users, who are either unwilling to take time to learn the tools or lacking in proper design expertise to create a professional infographic. In this paper, we explore an alternative approach: to automatically generate infographics from natural language statements. We first conducted a preliminary study to explore the design space of infographics. Based on the preliminary study, we built a proof-of-concept system that automatically converts statements about simple proportionrelated statistics to a set of infographics with pre-designed styles. Finally, we demonstrated the usability and usefulness of the system through sample results, exhibits, and expert reviews.
- Open Access
- 1 January 2018
- Published by Illuminating Engineering Society of Japan in Journal of Science and Technology in Lighting
- Vol. 41, 123-128
A Simple High-Sensitivity Acoustic Resonance Detection Method for Metal Halide Lamps
- 1 January 2018
- Published by Illuminating Engineering Society of Japan in Journal of Science and Technology in Lighting
- Vol. 41, 123-128
Open AccessA Simple High-Sensitivity Acoustic Resonance Detection Method for Metal Halide Lamps
- 8 July 2015
- Published by Springer Nature in Journal of Computer Science and Technology
- Vol. 30 (4) , 903-916
Leveraging Large Data with Weak Supervision for Joint Feature and Opinion Word Extraction
- 8 July 2015
- Published by Springer Nature in Journal of Computer Science and Technology
- Vol. 30 (4) , 903-916
Leveraging Large Data with Weak Supervision for Joint Feature and Opinion Word Extraction
- 22 May 2015
- Published by MyJove Corporation in Journal of Visualized Experiments
- No. 99,p. e52763
Comet Assay as an Indirect Measure of Systemic Oxidative Stress
Higher eukaryotic organisms cannot live without oxygen; yet, paradoxically, oxygen can be harmful to them. The oxygen molecule is chemically relatively inert because it has two unpaired electrons located in different pi * anti-bonding orbitals. These two electrons have parallel spins, meaning they rotate in the same direction about their own axes. This is why the oxygen molecule is not very reactive. Activation of oxygen may occur by two different mechanisms; either through reduction via one electron at a time (monovalent reduction), or through the absorption of sufficient energy to reverse the spin of one of the unpaired electrons. This results in the production of reactive oxidative species (ROS). There are a number of ways in which the human body eliminates ROS in its physiological state. If ROS production exceeds the repair capacity, oxidative stress results and damages different molecules. There are many different methods by which oxidative stress can be measured. This manuscript focuses on one of the methods named cell gel electrophoresis, also known as “comet assay” which allows measurement of DNA breaks. If all factors known to cause DNA damage, other than oxidative stress are kept constant, the amount of DNA damage measured by comet assay is a good parameter of oxidative stress. The principle is simple and relies upon the fact that DNA molecules are negatively charged. An intact DNA molecule has such a large size that it does not migrate during electrophoresis. DNA breaks, however, if present result in smaller fragments which move in the electrical field towards the anode. Smaller fragments migrate faster. As the fragments have different sizes the final result of the electrophoresis is not a distinct line but rather a continuum with the shape of a comet. The system allows a quantification of the resulting “comet” and thus of the DNA breaks in the cell.
- 22 May 2015
- Published by MyJove Corporation in Journal of Visualized Experiments
- No. 99,p. e52763
Comet Assay as an Indirect Measure of Systemic Oxidative Stress
Higher eukaryotic organisms cannot live without oxygen; yet, paradoxically, oxygen can be harmful to them. The oxygen molecule is chemically relatively inert because it has two unpaired electrons located in different pi * anti-bonding orbitals. These two electrons have parallel spins, meaning they rotate in the same direction about their own axes. This is why the oxygen molecule is not very reactive. Activation of oxygen may occur by two different mechanisms; either through reduction via one electron at a time (monovalent reduction), or through the absorption of sufficient energy to reverse the spin of one of the unpaired electrons. This results in the production of reactive oxidative species (ROS). There are a number of ways in which the human body eliminates ROS in its physiological state. If ROS production exceeds the repair capacity, oxidative stress results and damages different molecules. There are many different methods by which oxidative stress can be measured. This manuscript focuses on one of the methods named cell gel electrophoresis, also known as “comet assay” which allows measurement of DNA breaks. If all factors known to cause DNA damage, other than oxidative stress are kept constant, the amount of DNA damage measured by comet assay is a good parameter of oxidative stress. The principle is simple and relies upon the fact that DNA molecules are negatively charged. An intact DNA molecule has such a large size that it does not migrate during electrophoresis. DNA breaks, however, if present result in smaller fragments which move in the electrical field towards the anode. Smaller fragments migrate faster. As the fragments have different sizes the final result of the electrophoresis is not a distinct line but rather a continuum with the shape of a comet. The system allows a quantification of the resulting “comet” and thus of the DNA breaks in the cell.