Eduardo Coelho-Cerqueira
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
Research Subjects
Eduardo Coelho-Cerqueira has a total of 45 co-authors. Here are the top co-authors by number of shared publications:
- Cristian Follmer
Laboratory of Biological Chemistry of Neurodegenerative Disorders, Department of Physical Chemistry, Institute of Chemistry, Federal University of Rio de Janeiro (UFRJ) 1 , Rio de Janeiro 21941-909, Brazil
9 shared publications - Anderson S. Pinheiro
Departamento de Bioquímica, Instituto de Química, Universidade Federal do Rio de Janeiro, Rio de Janeiro 21941-902, RJ, Brazil
3 shared publications - Juliana Reis Cortines
Departamento de Virologia, Instituto de Microbiologia Paulo de Góes, Universidade Federal do Rio de Janeiro
2 shared publications - Paulo Augusto Netz
Instituto de Química da Universidade Federal do Rio Grande do Sul, UFRGS 7 , 91501-970 Porto Alegre, Rio Grande do Sul
2 shared publications - Allan Mariano de Souza
UNICAMP
1 shared publication Angelo C Pinto
Department of Organic Chemistry Institute of Chemistry Federal University of Rio de Janeiro Rio de Janeiro 21941‐909 Brazil
1 shared publication
Latest Publications
- 25 May 2026
- Published by Sociedade Brasileira de Computação
- p. 43-56
Agente K-alibra: Estratégia para Seleção de K-Clientes em Aprendizado Federado autônoma
O Aprendizado Federado (FL) estabeleceu-se como uma abordagem robusta para o treinamento distribuído de modelos, preservando a privacidade dos dados. Contudo, FL ainda sofre de problemas de escalabilidade e sobrecarga de rede. Uma das estratégias de mitigação pode ser através da seleção dinâmica de clientes, porém, a maioria dos trabalhos foca qualitativamente em quais clientes selecionar e não considera o número de clientes selecionados. Essa rigidez impede que o sistema se adapte adequadamente à dinâmica temporal do sistema distribuído, resultando em altos números de clientes ou poucos clientes de modo ineficiente. Essa falha é ocasionada pela dificuldade de algoritmos estáticos em adaptar-se dinamicamente às necessidades do sistema. Para mitigar essa rigidez, apresentamos o K-Agent, um orquestrador baseado em Language Models (LM) que ajusta dinamicamente o número de clientes via um ciclo de percepção, raciocínio e ação. Experimentos demonstram que o agente equilibra custos e estabilidade, reduzindo o tráfego de dados entre 44,4% e 59% comparado à literatura, validando a eficácia de agentes na otimização de hiperparâmetros em FL.
- 25 May 2026
- Published by Sociedade Brasileira de Computação
- p. 43-56
Agente K-alibra: Estratégia para Seleção de K-Clientes em Aprendizado Federado autônoma
O Aprendizado Federado (FL) estabeleceu-se como uma abordagem robusta para o treinamento distribuído de modelos, preservando a privacidade dos dados. Contudo, FL ainda sofre de problemas de escalabilidade e sobrecarga de rede. Uma das estratégias de mitigação pode ser através da seleção dinâmica de clientes, porém, a maioria dos trabalhos foca qualitativamente em quais clientes selecionar e não considera o número de clientes selecionados. Essa rigidez impede que o sistema se adapte adequadamente à dinâmica temporal do sistema distribuído, resultando em altos números de clientes ou poucos clientes de modo ineficiente. Essa falha é ocasionada pela dificuldade de algoritmos estáticos em adaptar-se dinamicamente às necessidades do sistema. Para mitigar essa rigidez, apresentamos o K-Agent, um orquestrador baseado em Language Models (LM) que ajusta dinamicamente o número de clientes via um ciclo de percepção, raciocínio e ação. Experimentos demonstram que o agente equilibra custos e estabilidade, reduzindo o tráfego de dados entre 44,4% e 59% comparado à literatura, validando a eficácia de agentes na otimização de hiperparâmetros em FL.
- Open Access
- 19 May 2025
- Published by Sociedade Brasileira de Computação
- p. 686-699
An Efficient Method for Model Compression for Federated Learning Scenarios
Traditional Machine Learning (ML) relies on centralized systems, requiring large amounts of sensitive data. In contrast, edge processing minimizes data transmission by performing computations locally, improving efficiency. In this context, Federated Learning (FL) is a decentralized machine learning approach that allows clients to train models collaboratively while keeping their data private. However, FL faces challenges such as high communication costs due to the large number of model updates between clients and the central server. In this paper, we introduce an efficient method for model compression for FL scenarios called HUFE-FL, which applies quantization to transform model parameters into discrete values, reducing their size, followed by Huffman encoding. This further compresses the model by efficiently encoding frequent symbols with fewer bits. Our main results show that HUFE-FL achieves up to 77.48% reduction in data transmission compared to traditional FL methods, with minimal loss in accuracy.
- 19 May 2025
- Published by Sociedade Brasileira de Computação
- p. 686-699
Open AccessAn Efficient Method for Model Compression for Federated Learning Scenarios
Traditional Machine Learning (ML) relies on centralized systems, requiring large amounts of sensitive data. In contrast, edge processing minimizes data transmission by performing computations locally, improving efficiency. In this context, Federated Learning (FL) is a decentralized machine learning approach that allows clients to train models collaboratively while keeping their data private. However, FL faces challenges such as high communication costs due to the large number of model updates between clients and the central server. In this paper, we introduce an efficient method for model compression for FL scenarios called HUFE-FL, which applies quantization to transform model parameters into discrete values, reducing their size, followed by Huffman encoding. This further compresses the model by efficiently encoding frequent symbols with fewer bits. Our main results show that HUFE-FL achieves up to 77.48% reduction in data transmission compared to traditional FL methods, with minimal loss in accuracy.
- Open Access
- 27 November 2023
- Published by Wiley in Mental Health Science
- Vol. 2 (1) , 57-72
The relationship between anxiety and COVID‐19‐associated isolation and lifestyle changes in the university community as a result of the SARS‐CoV‐2 pandemic
In 2020, the novel severe acute respiratory syndrome coronavirus outbreak emerged as a public health emergency. Three years later, we are still assessing the effects caused by the coronavirus disease (COVID‐19) and by the lifestyle changes due to prevention measures, like social isolation. Here, we assessed the profile of both the anxiety screening scores and anthropometric variables in the university community, associated with the isolation due to the COVID‐19 pandemic. We analyzed social and epidemiological associated factors for anxiety and weight change to address and better identify the physical and mental conditions aggravated during the pandemic. This is a cross‐sectional study based on the 7‐item Generalized Anxiety Disorder Scale questionnaire applied virtually to university students and professors in the State of Rio de Janeiro (n = 480). In our population cohort, we showed that the majority of individuals showed moderate to severe anxiety and increased body weight during the COVID‐19 pandemic. The major demographic associated factors for higher anxiety levels were young age, female gender, and low family income. The main modifiable associated factors for both anxiety and weight gain were the negative changes in living habits. In addition, there was a trend toward severe anxiety and greater weight change in people with increased time in social isolation. In view of these results, we suggest that bad habits, adopted during the COVID‐19 pandemic, were the main reason for the high anxiety levels found in the studied population.
- 27 November 2023
- Published by Wiley in Mental Health Science
- Vol. 2 (1) , 57-72
Open AccessThe relationship between anxiety and COVID‐19‐associated isolation and lifestyle changes in the university community as a result of the SARS‐CoV‐2 pandemic
In 2020, the novel severe acute respiratory syndrome coronavirus outbreak emerged as a public health emergency. Three years later, we are still assessing the effects caused by the coronavirus disease (COVID‐19) and by the lifestyle changes due to prevention measures, like social isolation. Here, we assessed the profile of both the anxiety screening scores and anthropometric variables in the university community, associated with the isolation due to the COVID‐19 pandemic. We analyzed social and epidemiological associated factors for anxiety and weight change to address and better identify the physical and mental conditions aggravated during the pandemic. This is a cross‐sectional study based on the 7‐item Generalized Anxiety Disorder Scale questionnaire applied virtually to university students and professors in the State of Rio de Janeiro (n = 480). In our population cohort, we showed that the majority of individuals showed moderate to severe anxiety and increased body weight during the COVID‐19 pandemic. The major demographic associated factors for higher anxiety levels were young age, female gender, and low family income. The main modifiable associated factors for both anxiety and weight gain were the negative changes in living habits. In addition, there was a trend toward severe anxiety and greater weight change in people with increased time in social isolation. In view of these results, we suggest that bad habits, adopted during the COVID‐19 pandemic, were the main reason for the high anxiety levels found in the studied population.
- 23 January 2020
- Published by American Chemical Society (ACS) in Journal of Chemical Education
- Vol. 97 (2) , 522-527
Determination of Zinc Oxide in Pharmaceutical Preparations by EDTA Titration: A Practical Class for a Quantitative Analysis Course
- 23 January 2020
- Published by American Chemical Society (ACS) in Journal of Chemical Education
- Vol. 97 (2) , 522-527
Determination of Zinc Oxide in Pharmaceutical Preparations by EDTA Titration: A Practical Class for a Quantitative Analysis Course
- 1 February 2019
- Published by Elsevier in Biochemical and Biophysical Research Communications
- Vol. 509 (2) , 367-372
Formation of large oligomers of DOPAL-modified α-synuclein is modulated by the oxidation of methionine residues located at C-terminal domain
- 1 February 2019
- Published by Elsevier in Biochemical and Biophysical Research Communications
- Vol. 509 (2) , 367-372
Formation of large oligomers of DOPAL-modified α-synuclein is modulated by the oxidation of methionine residues located at C-terminal domain
- 1 December 2018
- Published by Elsevier in Biochimica et Biophysica Acta (BBA) - General Subjects
- Vol. 1862 (12) , 2835-2845
In vitro neurotoxicity of salsolinol is attenuated by the presynaptic protein α-synuclein
- 1 December 2018
- Published by Elsevier in Biochimica et Biophysica Acta (BBA) - General Subjects
- Vol. 1862 (12) , 2835-2845
In vitro neurotoxicity of salsolinol is attenuated by the presynaptic protein α-synuclein
- 27 December 2016
- Published by Bentham Science Publishers Ltd. in Current Topics in Medicinal Chemistry
- Vol. 17 (4) , 489-497
A Medical Approach to the Monoamine Oxidase Inhibition by Using 7Hbenzo[ e]perimidin-7-one Derivatives
A series of perimidinone derivatives (7H-benzo[e]perimidin-7-one) were synthesized and assessed by means of in vitro assays as human MAO inhibitors. These compounds inhibited reversibly the enzymes with inhibitory constants in the range of 2 to 20 μM. In addition, the selectivity of inhibition of the MAO isoforms seems to be significantly dependent on the presence either of heteroatom or electron donating and withdrawing groups on the perimidinone framework, which was verified by using molecular docking simulation with the crystallized MAO receptors. Most of these inhibitors were highly selective: 9 and 11 inhibited selectively the MAO-B isoform while 12 had 10-fold selectivity for MAO-A isoform. Moreover, the compound 12 was both the most selective and potent MAO-A inhibitor among perimidinones. These results have important implications for the drug design of molecules targeting depression and movement-related disorders.
- 27 December 2016
- Published by Bentham Science Publishers Ltd. in Current Topics in Medicinal Chemistry
- Vol. 17 (4) , 489-497
A Medical Approach to the Monoamine Oxidase Inhibition by Using 7Hbenzo[ e]perimidin-7-one Derivatives
A series of perimidinone derivatives (7H-benzo[e]perimidin-7-one) were synthesized and assessed by means of in vitro assays as human MAO inhibitors. These compounds inhibited reversibly the enzymes with inhibitory constants in the range of 2 to 20 μM. In addition, the selectivity of inhibition of the MAO isoforms seems to be significantly dependent on the presence either of heteroatom or electron donating and withdrawing groups on the perimidinone framework, which was verified by using molecular docking simulation with the crystallized MAO receptors. Most of these inhibitors were highly selective: 9 and 11 inhibited selectively the MAO-B isoform while 12 had 10-fold selectivity for MAO-A isoform. Moreover, the compound 12 was both the most selective and potent MAO-A inhibitor among perimidinones. These results have important implications for the drug design of molecules targeting depression and movement-related disorders.
- Open Access
- 1 November 2015
- Published by Elsevier in Journal of Biological Chemistry
- Vol. 290 (46) , 27660-27679
Oligomerization and Membrane-binding Properties of Covalent Adducts Formed by the Interaction of α-Synuclein with the Toxic Dopamine Metabolite 3,4-Dihydroxyphenylacetaldehyde (DOPAL)
- 1 November 2015
- Published by Elsevier in Journal of Biological Chemistry
- Vol. 290 (46) , 27660-27679
Open AccessOligomerization and Membrane-binding Properties of Covalent Adducts Formed by the Interaction of α-Synuclein with the Toxic Dopamine Metabolite 3,4-Dihydroxyphenylacetaldehyde (DOPAL)
- 1 July 2014
- Published by Elsevier in Bioorganic & Medicinal Chemistry Letters
- Vol. 24 (14) , 3194-3198
Pitfalls associated with the use of Thioflavin-T to monitor anti-fibrillogenic activity
- 1 July 2014
- Published by Elsevier in Bioorganic & Medicinal Chemistry Letters
- Vol. 24 (14) , 3194-3198
Pitfalls associated with the use of Thioflavin-T to monitor anti-fibrillogenic activity
- 18 March 2014
- Published by Wiley in Chemical Biology & Drug Design
- Vol. 83 (4) , 401-410
Beyond Topoisomerase Inhibition: Antitumor 1,4‐Naphthoquinones as Potential Inhibitors of Human Monoamine Oxidase
Monoamine oxidase (MAO) action has been involved in the regulation of neurotransmitters levels, cell signaling, cellular growth, and differentiation as well as in the balance of the intracellular polyamine levels. Although so far obscure,MAOinhibitors are believed to have some effect on tumors progression. 1,4‐naphthoquinone (1,4‐NQ) has been pointed out as a potential pharmacophore for inhibition of bothMAOandDNAtopoisomerase activities, this latter associated with antitumor activity. Herein, we demonstrated that certain antitumor 1,4‐NQs, including spermidine‐1,4‐NQ, lapachol, andnor‐lapachol display inhibitory activity on humanMAO‐A andMAO‐B. Kinetic studies indicated that these compounds are reversible and competitiveMAOinhibitors, being the enzyme selectivity greatly affected by substitutions on 1,4‐NQring. Molecular docking studies suggested that the most potentMAOinhibitors are capable to bind to theMAOactive site in close proximity of flavin moiety. Furthermore, ability to inhibit bothMAO‐A andMAO‐B can be potentialized by the formation of hydrogen bonds between these compounds andFADand/or the residues in the active site. Although spermidine‐1,4‐NQs exhibit antitumor action primarily by inhibiting topoisomerase viaDNAintercalation, our findings suggest that their effect onMAOactivity should be taken into account when their application in cancer therapy is considered.
- 18 March 2014
- Published by Wiley in Chemical Biology & Drug Design
- Vol. 83 (4) , 401-410
Beyond Topoisomerase Inhibition: Antitumor 1,4‐Naphthoquinones as Potential Inhibitors of Human Monoamine Oxidase
Monoamine oxidase (MAO) action has been involved in the regulation of neurotransmitters levels, cell signaling, cellular growth, and differentiation as well as in the balance of the intracellular polyamine levels. Although so far obscure,MAOinhibitors are believed to have some effect on tumors progression. 1,4‐naphthoquinone (1,4‐NQ) has been pointed out as a potential pharmacophore for inhibition of bothMAOandDNAtopoisomerase activities, this latter associated with antitumor activity. Herein, we demonstrated that certain antitumor 1,4‐NQs, including spermidine‐1,4‐NQ, lapachol, andnor‐lapachol display inhibitory activity on humanMAO‐A andMAO‐B. Kinetic studies indicated that these compounds are reversible and competitiveMAOinhibitors, being the enzyme selectivity greatly affected by substitutions on 1,4‐NQring. Molecular docking studies suggested that the most potentMAOinhibitors are capable to bind to theMAOactive site in close proximity of flavin moiety. Furthermore, ability to inhibit bothMAO‐A andMAO‐B can be potentialized by the formation of hydrogen bonds between these compounds andFADand/or the residues in the active site. Although spermidine‐1,4‐NQs exhibit antitumor action primarily by inhibiting topoisomerase viaDNAintercalation, our findings suggest that their effect onMAOactivity should be taken into account when their application in cancer therapy is considered.