Andre Python
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
Andre Python has a total of 108 co-authors. Here are the top co-authors by number of shared publications:
- Timothy C. D. Lucas
University of Leicester
5 shared publications - Jianwei Yin
School of Software Technology, Zhejiang University, Ningbo, China
4 shared publications - Jürgen Brandsch
Bonn International Center for Conversion (BICC)
4 shared publications - Andreas Bender
Department of Statistics, LMU Munich, Munich, Germany
3 shared publications - Baoli Liu
Xiamen University
3 shared publications - HongYi Li
Center for Data Science, Zhejiang University, Hangzhou, China
3 shared publications
Latest Publications
- Open Access
Beyond straight lines: migration costs considering geography enhance tracing human genetic ancestry
Reconstructing the spatio-temporal history of human genetic lineages is fundamental to understanding human evolution and population distribution. While succinct tree sequences and maximum parsimony reconstruction methods applied to large-scale genomic data have improved our ability to trace the geographic history of genetic ancestry, they have essentially relied on Euclidean distances, which ineluctably ignore opportunity costs that have shaped human mobility patterns since the earliest human migrations and settlement formations. Here we propose an approach to incorporate realistic geographical migration costs through a human movement friction surface. Using simulated data mimicking the dispersal process of human migration out of Africa, we found that, compared to the Euclidean-based benchmark (M 0 ), the proposed friction-based model (M f ) leads to a more accurate estimation of the geographical origin ( n = 346, accuracy M 0 = 0.18, f = 0.27) and genetic flux ( n = 30, MSE M 0 = 0.20, M f = 0.12) through the Mandeb corridor in the Horn of Africa. We further illustrate these findings in a case study, in which our model seems to better identify plausible human migration paths from Eurasia to the Americas by accounting for geographic factors affecting migration opportunity costs, such as the Alaska Range and Rocky Mountains that represent physical barriers that constraint migration. While important migration drivers such as climate change, technological advances, social organization, and culture remain omitted here, our work highlights the importance of explicitly accounting for geographic constraints to improve our ability to reconstruct past human mobility and, ultimately, understand the evolution of human populations.
Open AccessBeyond straight lines: migration costs considering geography enhance tracing human genetic ancestry
Reconstructing the spatio-temporal history of human genetic lineages is fundamental to understanding human evolution and population distribution. While succinct tree sequences and maximum parsimony reconstruction methods applied to large-scale genomic data have improved our ability to trace the geographic history of genetic ancestry, they have essentially relied on Euclidean distances, which ineluctably ignore opportunity costs that have shaped human mobility patterns since the earliest human migrations and settlement formations. Here we propose an approach to incorporate realistic geographical migration costs through a human movement friction surface. Using simulated data mimicking the dispersal process of human migration out of Africa, we found that, compared to the Euclidean-based benchmark (M 0 ), the proposed friction-based model (M f ) leads to a more accurate estimation of the geographical origin ( n = 346, accuracy M 0 = 0.18, f = 0.27) and genetic flux ( n = 30, MSE M 0 = 0.20, M f = 0.12) through the Mandeb corridor in the Horn of Africa. We further illustrate these findings in a case study, in which our model seems to better identify plausible human migration paths from Eurasia to the Americas by accounting for geographic factors affecting migration opportunity costs, such as the Alaska Range and Rocky Mountains that represent physical barriers that constraint migration. While important migration drivers such as climate change, technological advances, social organization, and culture remain omitted here, our work highlights the importance of explicitly accounting for geographic constraints to improve our ability to reconstruct past human mobility and, ultimately, understand the evolution of human populations.
- Open Access
- 24 September 2025
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 27 (1) , e82729
Author's Reply: Critical Limitations in Systematic Reviews of Large Language Models in Health Care
- 24 September 2025
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 27 (1) , e82729
Open AccessAuthor's Reply: Critical Limitations in Systematic Reviews of Large Language Models in Health Care
- Open Access
Geographical variation drives adaptive equilibrium of the P. falciparum sickle-associated mutations
The recent discovery of genetic mutations in Plasmodium falciparum —the most lethal malaria parasite—that enable it to overcome the protective effects of sickle cell trait, raises fundamental questions about the underlying biological and evolutionary interactions. Here we develop a geostatistical model to compare sickle haemoglobin genotype frequencies to the Plasmodium falciparum sickle-associated alleles across global populations, and find a robust association at multiple geographical scales, implying that sickle drives positive selection for these parasite mutations. A model of parasite evolution and an analysis of local haplotype patterns suggest that key features of these mutations – that they are polymorphic in all African populations and are mutually correlated despite lying in different genome regions - are caused by geographical variation in selection pressure, and that the alleles may have been maintained by balancing selection over timescales comparable to the age of the sickle mutation itself. The predicted impact of this host-parasite interaction on disease outcomes varies widely across populations, and functional data are needed to discover the biological mechanisms involved.
Open AccessGeographical variation drives adaptive equilibrium of the P. falciparum sickle-associated mutations
The recent discovery of genetic mutations in Plasmodium falciparum —the most lethal malaria parasite—that enable it to overcome the protective effects of sickle cell trait, raises fundamental questions about the underlying biological and evolutionary interactions. Here we develop a geostatistical model to compare sickle haemoglobin genotype frequencies to the Plasmodium falciparum sickle-associated alleles across global populations, and find a robust association at multiple geographical scales, implying that sickle drives positive selection for these parasite mutations. A model of parasite evolution and an analysis of local haplotype patterns suggest that key features of these mutations – that they are polymorphic in all African populations and are mutually correlated despite lying in different genome regions - are caused by geographical variation in selection pressure, and that the alleles may have been maintained by balancing selection over timescales comparable to the age of the sickle mutation itself. The predicted impact of this host-parasite interaction on disease outcomes varies widely across populations, and functional data are needed to discover the biological mechanisms involved.
- Top Cited PapersOpen Access
- 11 July 2025
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 27 (1) , e71916
Implementing Large Language Models in Health Care: Clinician-Focused Review With Interactive Guideline
Background: Large language models (LLMs) can generate outputs understandable by humans, such as answers to medical questions and radiology reports. With the rapid development of LLMs, clinicians face a growing challenge in determining the most suitable algorithms to support their work. Objective: We aimed to provide clinicians and other health care practitioners with systematic guidance in selecting an LLM that is relevant and appropriate to their needs and facilitate the integration process of LLMs in health care. Methods: We conducted a literature search of full-text publications in English on clinical applications of LLMs published between January 1, 2022, and March 31, 2025, on PubMed, ScienceDirect, Scopus, and IEEE Xplore. We excluded papers from journals below a set citation threshold, as well as papers that did not focus on LLMs, were not research based, or did not involve clinical applications. We also conducted a literature search on arXiv within the same investigated period and included papers on the clinical applications of innovative multimodal LLMs. This led to a total of 270 studies. Results: We collected 330 LLMs and recorded their application frequency in clinical tasks and frequency of best performance in their context. On the basis of a 5-stage clinical workflow, we found that stages 2, 3, and 4 are key stages in the clinical workflow, involving numerous clinical subtasks and LLMs. However, the diversity of LLMs that may perform optimally in each context remains limited. GPT-3.5 and GPT-4 were the most versatile models in the 5-stage clinical workflow, applied to 52% (29/56) and 71% (40/56) of the clinical subtasks, respectively, and they performed best in 29% (16/56) and 54% (30/56) of the clinical subtasks, respectively. General-purpose LLMs may not perform well in specialized areas as they often require lightweight prompt engineering methods or fine-tuning techniques based on specific datasets to improve model performance. Most LLMs with multimodal abilities are closed-source models and, therefore, lack of transparency, model customization, and fine-tuning for specific clinical tasks and may also pose challenges regarding data protection and privacy, which are common requirements in clinical settings. Conclusions: In this review, we found that LLMs may help clinicians in a variety of clinical tasks. However, we did not find evidence of generalist clinical LLMs successfully applicable to a wide range of clinical tasks. Therefore, their clinical deployment remains challenging. On the basis of this review, we propose an interactive online guideline for clinicians to select suitable LLMs by clinical task. With a clinical perspective and free of unnecessary technical jargon, this guideline may be used as a reference to successfully apply LLMs in clinical settings.
- 11 July 2025
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 27 (1) , e71916
Top Cited PapersOpen AccessImplementing Large Language Models in Health Care: Clinician-Focused Review With Interactive Guideline
Background: Large language models (LLMs) can generate outputs understandable by humans, such as answers to medical questions and radiology reports. With the rapid development of LLMs, clinicians face a growing challenge in determining the most suitable algorithms to support their work. Objective: We aimed to provide clinicians and other health care practitioners with systematic guidance in selecting an LLM that is relevant and appropriate to their needs and facilitate the integration process of LLMs in health care. Methods: We conducted a literature search of full-text publications in English on clinical applications of LLMs published between January 1, 2022, and March 31, 2025, on PubMed, ScienceDirect, Scopus, and IEEE Xplore. We excluded papers from journals below a set citation threshold, as well as papers that did not focus on LLMs, were not research based, or did not involve clinical applications. We also conducted a literature search on arXiv within the same investigated period and included papers on the clinical applications of innovative multimodal LLMs. This led to a total of 270 studies. Results: We collected 330 LLMs and recorded their application frequency in clinical tasks and frequency of best performance in their context. On the basis of a 5-stage clinical workflow, we found that stages 2, 3, and 4 are key stages in the clinical workflow, involving numerous clinical subtasks and LLMs. However, the diversity of LLMs that may perform optimally in each context remains limited. GPT-3.5 and GPT-4 were the most versatile models in the 5-stage clinical workflow, applied to 52% (29/56) and 71% (40/56) of the clinical subtasks, respectively, and they performed best in 29% (16/56) and 54% (30/56) of the clinical subtasks, respectively. General-purpose LLMs may not perform well in specialized areas as they often require lightweight prompt engineering methods or fine-tuning techniques based on specific datasets to improve model performance. Most LLMs with multimodal abilities are closed-source models and, therefore, lack of transparency, model customization, and fine-tuning for specific clinical tasks and may also pose challenges regarding data protection and privacy, which are common requirements in clinical settings. Conclusions: In this review, we found that LLMs may help clinicians in a variety of clinical tasks. However, we did not find evidence of generalist clinical LLMs successfully applicable to a wide range of clinical tasks. Therefore, their clinical deployment remains challenging. On the basis of this review, we propose an interactive online guideline for clinicians to select suitable LLMs by clinical task. With a clinical perspective and free of unnecessary technical jargon, this guideline may be used as a reference to successfully apply LLMs in clinical settings.
- Open Access
- 15 March 2025
- Published by Copernicus GmbH
GeoDaedalus: Automatic Geoscience Dataset Construction
Many subfields of Geosciences are currently experiencing the long-tail data distribution curse. While there are large head databases within the fields, they are updated slowly, are few in number, and have poor interoperability between the few existing ones. More often, data is generated by research groups through experiments, combined with other data collected on the same topic to form a small dataset, which is hidden in the scientific literature. These chaotic data organizing manners result in low utilization rates of new data in the scientific community, hindering the implementation of data FAIR principles. To contribute to improve the process chain of long-tail data collection and linking in science, we propose GeoDaedalus, a multi-agenic large language models (LLM)-based architecture for on-demand automatic geoscience dataset construction. Starting from the research needs, GeoDaedalus achieves end-to-end automation of the scientific data curation process through a series of processes, including online search, information matching & extraction, and data fusion. To access the efficiency and accuracy in data extraction in GeoDaedalus, we simulated different use cases such as those in Geochemistry, along with complete human expert data collection processes, and constructed the first benchmark for evaluating scientific data curation processes: GeoDataBench. Results from the latest multimodal LLMs to evaluate GeoDaedalus on GeoDataBench suggest better capabilities with lower economic costs, which may become a new benchmark for GeoDataBench. We propose a Python API package with an interpretable full-process transparent logging module suitable for GeoDaedalus' users to address the highly customized needs of scientific work. Although GeoDaedalus uses geoscience data as a sample, its relevant capabilities, once reorganized, can extend to other scientific fields, marking a solid step towards Open Science for the scientific community.
- 15 March 2025
- Published by Copernicus GmbH
Open AccessGeoDaedalus: Automatic Geoscience Dataset Construction
Many subfields of Geosciences are currently experiencing the long-tail data distribution curse. While there are large head databases within the fields, they are updated slowly, are few in number, and have poor interoperability between the few existing ones. More often, data is generated by research groups through experiments, combined with other data collected on the same topic to form a small dataset, which is hidden in the scientific literature. These chaotic data organizing manners result in low utilization rates of new data in the scientific community, hindering the implementation of data FAIR principles. To contribute to improve the process chain of long-tail data collection and linking in science, we propose GeoDaedalus, a multi-agenic large language models (LLM)-based architecture for on-demand automatic geoscience dataset construction. Starting from the research needs, GeoDaedalus achieves end-to-end automation of the scientific data curation process through a series of processes, including online search, information matching & extraction, and data fusion. To access the efficiency and accuracy in data extraction in GeoDaedalus, we simulated different use cases such as those in Geochemistry, along with complete human expert data collection processes, and constructed the first benchmark for evaluating scientific data curation processes: GeoDataBench. Results from the latest multimodal LLMs to evaluate GeoDaedalus on GeoDataBench suggest better capabilities with lower economic costs, which may become a new benchmark for GeoDataBench. We propose a Python API package with an interpretable full-process transparent logging module suitable for GeoDaedalus' users to address the highly customized needs of scientific work. Although GeoDaedalus uses geoscience data as a sample, its relevant capabilities, once reorganized, can extend to other scientific fields, marking a solid step towards Open Science for the scientific community.
- Open Access
- 29 January 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Implementing Large Language Models in Health Care: Clinician-Focused Review With Interactive Guideline (Preprint)
BACKGROUND Large language models (LLMs) can generate outputs understandable by humans, such as answers to medical questions and radiology reports. With the rapid development of LLMs, clinicians face a growing challenge in determining the most suitable algorithms to support their work. OBJECTIVE We aimed to provide clinicians and other health care practitioners with systematic guidance in selecting an LLM that is relevant and appropriate to their needs and facilitate the integration process of LLMs in health care. METHODS We conducted a literature search of full-text publications in English on clinical applications of LLMs published between January 1, 2022, and March 31, 2025, on PubMed, ScienceDirect, Scopus, and IEEE Xplore. We excluded papers from journals below a set citation threshold, as well as papers that did not focus on LLMs, were not research based, or did not involve clinical applications. We also conducted a literature search on arXiv within the same investigated period and included papers on the clinical applications of innovative multimodal LLMs. This led to a total of 270 studies. RESULTS We collected 330 LLMs and recorded their application frequency in clinical tasks and frequency of best performance in their context. On the basis of a 5-stage clinical workflow, we found that stages 2, 3, and 4 are key stages in the clinical workflow, involving numerous clinical subtasks and LLMs. However, the diversity of LLMs that may perform optimally in each context remains limited. GPT-3.5 and GPT-4 were the most versatile models in the 5-stage clinical workflow, applied to 52% (29/56) and 71% (40/56) of the clinical subtasks, respectively, and they performed best in 29% (16/56) and 54% (30/56) of the clinical subtasks, respectively. General-purpose LLMs may not perform well in specialized areas as they often require lightweight prompt engineering methods or fine-tuning techniques based on specific datasets to improve model performance. Most LLMs with multimodal abilities are closed-source models and, therefore, lack of transparency, model customization, and fine-tuning for specific clinical tasks and may also pose challenges regarding data protection and privacy, which are common requirements in clinical settings. CONCLUSIONS In this review, we found that LLMs may help clinicians in a variety of clinical tasks. However, we did not find evidence of generalist clinical LLMs successfully applicable to a wide range of clinical tasks. Therefore, their clinical deployment remains challenging. On the basis of this review, we propose an interactive online guideline for clinicians to select suitable LLMs by clinical task. With a clinical perspective and free of unnecessary technical jargon, this guideline may be used as a reference to successfully apply LLMs in clinical settings.
- 29 January 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessImplementing Large Language Models in Health Care: Clinician-Focused Review With Interactive Guideline (Preprint)
BACKGROUND Large language models (LLMs) can generate outputs understandable by humans, such as answers to medical questions and radiology reports. With the rapid development of LLMs, clinicians face a growing challenge in determining the most suitable algorithms to support their work. OBJECTIVE We aimed to provide clinicians and other health care practitioners with systematic guidance in selecting an LLM that is relevant and appropriate to their needs and facilitate the integration process of LLMs in health care. METHODS We conducted a literature search of full-text publications in English on clinical applications of LLMs published between January 1, 2022, and March 31, 2025, on PubMed, ScienceDirect, Scopus, and IEEE Xplore. We excluded papers from journals below a set citation threshold, as well as papers that did not focus on LLMs, were not research based, or did not involve clinical applications. We also conducted a literature search on arXiv within the same investigated period and included papers on the clinical applications of innovative multimodal LLMs. This led to a total of 270 studies. RESULTS We collected 330 LLMs and recorded their application frequency in clinical tasks and frequency of best performance in their context. On the basis of a 5-stage clinical workflow, we found that stages 2, 3, and 4 are key stages in the clinical workflow, involving numerous clinical subtasks and LLMs. However, the diversity of LLMs that may perform optimally in each context remains limited. GPT-3.5 and GPT-4 were the most versatile models in the 5-stage clinical workflow, applied to 52% (29/56) and 71% (40/56) of the clinical subtasks, respectively, and they performed best in 29% (16/56) and 54% (30/56) of the clinical subtasks, respectively. General-purpose LLMs may not perform well in specialized areas as they often require lightweight prompt engineering methods or fine-tuning techniques based on specific datasets to improve model performance. Most LLMs with multimodal abilities are closed-source models and, therefore, lack of transparency, model customization, and fine-tuning for specific clinical tasks and may also pose challenges regarding data protection and privacy, which are common requirements in clinical settings. CONCLUSIONS In this review, we found that LLMs may help clinicians in a variety of clinical tasks. However, we did not find evidence of generalist clinical LLMs successfully applicable to a wide range of clinical tasks. Therefore, their clinical deployment remains challenging. On the basis of this review, we propose an interactive online guideline for clinicians to select suitable LLMs by clinical task. With a clinical perspective and free of unnecessary technical jargon, this guideline may be used as a reference to successfully apply LLMs in clinical settings.
- Open Access
- 28 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 494-508
QuCT: A Framework for Analyzing Quantum Circuit by Extracting Contextual and Topological Features
In the current Noisy Intermediate-Scale Quantum era, quantum circuit analysis is an essential technique for designing high-performance quantum programs. Current analysis methods exhibit either accuracy limitations or high computational complexity for obtaining precise results. To reduce this tradeoff, we propose QuCT, a unified framework for extracting, analyzing, and optimizing quantum circuits. The main innovation of QuCT is to vectorize each gate with each element, quantitatively describing the degree of the interaction with neighboring gates. Extending from the vectorization model, we propose two representative downstream models for fidelity prediction and unitary decomposition. The fidelity prediction model performs a linear transformation on all gate vectors and aggregates the results to estimate the overall circuit fidelity. By identifying critical weights in the transformation matrix, we propose two optimizations to improve the circuit fidelity. In the unitary decomposition model, we significantly reduce the search space by bridging the gap between unitary and circuit via gate vectors. Experiments show that QuCT improves the accuracy of fidelity prediction by 4.2 × on 5-qubit and 18-qubit quantum devices and achieves 2.5 × fidelity improvement compared to existing quantum compilers [19, 55]. In unitary decomposition, QuCT achieves 46.3 × speedup for 5-qubit unitary and more than hundreds of speedup for 8-qubit unitary, compared to the state-of-the-art method [87].
- 28 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 494-508
Open AccessQuCT: A Framework for Analyzing Quantum Circuit by Extracting Contextual and Topological Features
In the current Noisy Intermediate-Scale Quantum era, quantum circuit analysis is an essential technique for designing high-performance quantum programs. Current analysis methods exhibit either accuracy limitations or high computational complexity for obtaining precise results. To reduce this tradeoff, we propose QuCT, a unified framework for extracting, analyzing, and optimizing quantum circuits. The main innovation of QuCT is to vectorize each gate with each element, quantitatively describing the degree of the interaction with neighboring gates. Extending from the vectorization model, we propose two representative downstream models for fidelity prediction and unitary decomposition. The fidelity prediction model performs a linear transformation on all gate vectors and aggregates the results to estimate the overall circuit fidelity. By identifying critical weights in the transformation matrix, we propose two optimizations to improve the circuit fidelity. In the unitary decomposition model, we significantly reduce the search space by bridging the gap between unitary and circuit via gate vectors. Experiments show that QuCT improves the accuracy of fidelity prediction by 4.2 × on 5-qubit and 18-qubit quantum devices and achieves 2.5 × fidelity improvement compared to existing quantum compilers [19, 55]. In unitary decomposition, QuCT achieves 46.3 × speedup for 5-qubit unitary and more than hundreds of speedup for 8-qubit unitary, compared to the state-of-the-art method [87].
- 13 October 2023
- Published by Taylor & Francis in International Journal of Geographical Information Science
- Vol. 37 (11) , 2319-2344
Extending regionalization algorithms to explore spatial process heterogeneity
- 13 October 2023
- Published by Taylor & Francis in International Journal of Geographical Information Science
- Vol. 37 (11) , 2319-2344
Extending regionalization algorithms to explore spatial process heterogeneity
- Open Access
- 4 July 2023
- Published by Springer Nature in Communications Physics
- Vol. 6 (1) , 1-11
Fast crystal growth of ice VII owing to the decoupling of translational and rotational ordering
Despite the abundance of water’s crystalline polymorphs, the growth mechanisms of most ice forms remain poorly understood. This study applies extensive molecular dynamics (MD) simulations to examine the growth of ice VII, revealing a fast growth rate comparable to pure metals while maintaining robust hydrogen-bond networks. The results from an unsupervised machine learning applied to identify local structure suggest that the surface of ice VII consistently exhibits a body-centered cubic (bcc) plastic ice layer, indicating the decoupling of translational and rotational orderings. The study also uncovers the ultrafast growth rate of pure plastic ice, indicating that orientational disorder in the crystal structure may be associated with faster kinetics. Additionally, we discuss the impacts of interfacial plastic layer width and surface roughness on growth mode.
- 4 July 2023
- Published by Springer Nature in Communications Physics
- Vol. 6 (1) , 1-11
Open AccessFast crystal growth of ice VII owing to the decoupling of translational and rotational ordering
Despite the abundance of water’s crystalline polymorphs, the growth mechanisms of most ice forms remain poorly understood. This study applies extensive molecular dynamics (MD) simulations to examine the growth of ice VII, revealing a fast growth rate comparable to pure metals while maintaining robust hydrogen-bond networks. The results from an unsupervised machine learning applied to identify local structure suggest that the surface of ice VII consistently exhibits a body-centered cubic (bcc) plastic ice layer, indicating the decoupling of translational and rotational orderings. The study also uncovers the ultrafast growth rate of pure plastic ice, indicating that orientational disorder in the crystal structure may be associated with faster kinetics. Additionally, we discuss the impacts of interfacial plastic layer width and surface roughness on growth mode.
- Open Access
- 13 February 2023
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series A: Statistics in Society
- Vol. 186 (2) , 194
Andre Python's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al.
Andre Python; Experimental Evaluation of Algorithm-Assisted Human Decision-Making: Application to Pretrial Public Safety Assessment, Journal of the Royal Statis
- 13 February 2023
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series A: Statistics in Society
- Vol. 186 (2) , 194
Open AccessAndre Python's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al.
Andre Python; Experimental Evaluation of Algorithm-Assisted Human Decision-Making: Application to Pretrial Public Safety Assessment, Journal of the Royal Statis