Human Genome Polymorphisms and Computational Intelligence Approach Revealed a Complex Genomic Signature for COVID-19 Severity in Brazilian Patients
We present a genome polymorphisms/machine learning approach for severe COVID-19 prognosis. Ninety-six Brazilian severe COVID-19 patients and controls were genotyped for 296 innate immunity loci. Our model used a feature selection algorithm, namely recursive feature elimination coupled with a support...
| Main Authors: | , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Viruses |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-4915/15/3/645 |
| _version_ | 1797608590034337792 |
|---|---|
| author | André Filipe Pastor Cássia Docena Antônio Mauro Rezende Flávio Rosendo da Silva Oliveira Marília de Albuquerque Sena Clarice Neuenschwander Lins de Morais Cristiane Campello Bresani-Salvi Luydson Richardson Silva Vasconcelos Kennya Danielle Campelo Valença Carolline de Araújo Mariz Carlos Brito Cláudio Duarte Fonseca Cynthia Braga Christian Robson de Souza Reis Ernesto Torres de Azevedo Marques Bartolomeu Acioli-Santos |
| author_facet | André Filipe Pastor Cássia Docena Antônio Mauro Rezende Flávio Rosendo da Silva Oliveira Marília de Albuquerque Sena Clarice Neuenschwander Lins de Morais Cristiane Campello Bresani-Salvi Luydson Richardson Silva Vasconcelos Kennya Danielle Campelo Valença Carolline de Araújo Mariz Carlos Brito Cláudio Duarte Fonseca Cynthia Braga Christian Robson de Souza Reis Ernesto Torres de Azevedo Marques Bartolomeu Acioli-Santos |
| author_sort | André Filipe Pastor |
| collection | DOAJ |
| description | We present a genome polymorphisms/machine learning approach for severe COVID-19 prognosis. Ninety-six Brazilian severe COVID-19 patients and controls were genotyped for 296 innate immunity loci. Our model used a feature selection algorithm, namely recursive feature elimination coupled with a support vector machine, to find the optimal loci classification subset, followed by a support vector machine with the linear kernel (SVM-LK) to classify patients into the severe COVID-19 group. The best features that were selected by the SVM-RFE method included 12 SNPs in 12 genes: <i>PD-L1</i>, <i>PD-L2</i>, <i>IL10RA</i>, <i>JAK2</i>, <i>STAT1</i>, <i>IFIT1</i>, <i>IFIH1</i>, <i>DC-SIGNR</i>, <i>IFNB1</i>, <i>IRAK4</i>, <i>IRF1</i>, and <i>IL10</i>. During the COVID-19 prognosis step by SVM-LK, the metrics were: 85% accuracy, 80% sensitivity, and 90% specificity. In comparison, univariate analysis under the 12 selected SNPs showed some highlights for individual variant alleles that represented risk (<i>PD-L1</i> and <i>IFIT1</i>) or protection (<i>JAK2</i> and <i>IFIH1</i>). Variant genotypes carrying risk effects were represented by <i>PD-L2</i> and <i>IFIT1</i> genes. The proposed complex classification method can be used to identify individuals who are at a high risk of developing severe COVID-19 outcomes even in uninfected conditions, which is a disruptive concept in COVID-19 prognosis. Our results suggest that the genetic context is an important factor in the development of severe COVID-19. |
| first_indexed | 2024-03-11T05:46:25Z |
| format | Article |
| id | doaj.art-06d9f92637ff49e9989139186e6e38e6 |
| institution | Directory Open Access Journal |
| issn | 1999-4915 |
| language | English |
| last_indexed | 2024-03-11T05:46:25Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Viruses |
| spelling | doaj.art-06d9f92637ff49e9989139186e6e38e62023-11-17T14:22:19ZengMDPI AGViruses1999-49152023-02-0115364510.3390/v15030645Human Genome Polymorphisms and Computational Intelligence Approach Revealed a Complex Genomic Signature for COVID-19 Severity in Brazilian PatientsAndré Filipe Pastor0Cássia Docena1Antônio Mauro Rezende2Flávio Rosendo da Silva Oliveira3Marília de Albuquerque Sena4Clarice Neuenschwander Lins de Morais5Cristiane Campello Bresani-Salvi6Luydson Richardson Silva Vasconcelos7Kennya Danielle Campelo Valença8Carolline de Araújo Mariz9Carlos Brito10Cláudio Duarte Fonseca11Cynthia Braga12Christian Robson de Souza Reis13Ernesto Torres de Azevedo Marques14Bartolomeu Acioli-Santos15Sertão Pernambucano Federal Institute of Education, Science and Technology, Petrolina 56316-686, PE, BrazilCore Facility, Oswaldo Cruz Foundation, Recife 50740-465, PE, BrazilDepartment of Microbiology, Aggeu Magalhães Institute, Oswaldo Cruz Foundation, Recife 50740-465, PE, BrazilFederal Institute of Education, Science and Technology of Pernambuco, Recife 50740-545, PE, BrazilDepartment of Virology, Aggeu Magalhães Institute, Oswaldo Cruz Foundation, Recife 50740-465, PE, BrazilDepartment of Virology, Aggeu Magalhães Institute, Oswaldo Cruz Foundation, Recife 50740-465, PE, BrazilDepartment of Virology, Aggeu Magalhães Institute, Oswaldo Cruz Foundation, Recife 50740-465, PE, BrazilDepartment of Parasitology, Aggeu Magalhães Institute, Oswaldo Cruz Foundation, Recife 50740-465, PE, BrazilDepartment of Virology, Aggeu Magalhães Institute, Oswaldo Cruz Foundation, Recife 50740-465, PE, BrazilDepartment of Parasitology, Aggeu Magalhães Institute, Oswaldo Cruz Foundation, Recife 50740-465, PE, BrazilDepartment of Clinical Medicine, Pernambuco Federal University, Recife 50740-600, PE, BrazilServidores do Estado Hospital (HSE), Recife 52020-020, PE, BrazilDepartment of Parasitology, Aggeu Magalhães Institute, Oswaldo Cruz Foundation, Recife 50740-465, PE, BrazilDepartment of Microbiology, Aggeu Magalhães Institute, Oswaldo Cruz Foundation, Recife 50740-465, PE, BrazilDepartment of Virology, Aggeu Magalhães Institute, Oswaldo Cruz Foundation, Recife 50740-465, PE, BrazilDepartment of Virology, Aggeu Magalhães Institute, Oswaldo Cruz Foundation, Recife 50740-465, PE, BrazilWe present a genome polymorphisms/machine learning approach for severe COVID-19 prognosis. Ninety-six Brazilian severe COVID-19 patients and controls were genotyped for 296 innate immunity loci. Our model used a feature selection algorithm, namely recursive feature elimination coupled with a support vector machine, to find the optimal loci classification subset, followed by a support vector machine with the linear kernel (SVM-LK) to classify patients into the severe COVID-19 group. The best features that were selected by the SVM-RFE method included 12 SNPs in 12 genes: <i>PD-L1</i>, <i>PD-L2</i>, <i>IL10RA</i>, <i>JAK2</i>, <i>STAT1</i>, <i>IFIT1</i>, <i>IFIH1</i>, <i>DC-SIGNR</i>, <i>IFNB1</i>, <i>IRAK4</i>, <i>IRF1</i>, and <i>IL10</i>. During the COVID-19 prognosis step by SVM-LK, the metrics were: 85% accuracy, 80% sensitivity, and 90% specificity. In comparison, univariate analysis under the 12 selected SNPs showed some highlights for individual variant alleles that represented risk (<i>PD-L1</i> and <i>IFIT1</i>) or protection (<i>JAK2</i> and <i>IFIH1</i>). Variant genotypes carrying risk effects were represented by <i>PD-L2</i> and <i>IFIT1</i> genes. The proposed complex classification method can be used to identify individuals who are at a high risk of developing severe COVID-19 outcomes even in uninfected conditions, which is a disruptive concept in COVID-19 prognosis. Our results suggest that the genetic context is an important factor in the development of severe COVID-19.https://www.mdpi.com/1999-4915/15/3/645COVID-19 geneticsSARS-CoV-2 infectioncomplex genomic classifiermachine learning |
| spellingShingle | André Filipe Pastor Cássia Docena Antônio Mauro Rezende Flávio Rosendo da Silva Oliveira Marília de Albuquerque Sena Clarice Neuenschwander Lins de Morais Cristiane Campello Bresani-Salvi Luydson Richardson Silva Vasconcelos Kennya Danielle Campelo Valença Carolline de Araújo Mariz Carlos Brito Cláudio Duarte Fonseca Cynthia Braga Christian Robson de Souza Reis Ernesto Torres de Azevedo Marques Bartolomeu Acioli-Santos Human Genome Polymorphisms and Computational Intelligence Approach Revealed a Complex Genomic Signature for COVID-19 Severity in Brazilian Patients Viruses COVID-19 genetics SARS-CoV-2 infection complex genomic classifier machine learning |
| title | Human Genome Polymorphisms and Computational Intelligence Approach Revealed a Complex Genomic Signature for COVID-19 Severity in Brazilian Patients |
| title_full | Human Genome Polymorphisms and Computational Intelligence Approach Revealed a Complex Genomic Signature for COVID-19 Severity in Brazilian Patients |
| title_fullStr | Human Genome Polymorphisms and Computational Intelligence Approach Revealed a Complex Genomic Signature for COVID-19 Severity in Brazilian Patients |
| title_full_unstemmed | Human Genome Polymorphisms and Computational Intelligence Approach Revealed a Complex Genomic Signature for COVID-19 Severity in Brazilian Patients |
| title_short | Human Genome Polymorphisms and Computational Intelligence Approach Revealed a Complex Genomic Signature for COVID-19 Severity in Brazilian Patients |
| title_sort | human genome polymorphisms and computational intelligence approach revealed a complex genomic signature for covid 19 severity in brazilian patients |
| topic | COVID-19 genetics SARS-CoV-2 infection complex genomic classifier machine learning |
| url | https://www.mdpi.com/1999-4915/15/3/645 |
| work_keys_str_mv | AT andrefilipepastor humangenomepolymorphismsandcomputationalintelligenceapproachrevealedacomplexgenomicsignatureforcovid19severityinbrazilianpatients AT cassiadocena humangenomepolymorphismsandcomputationalintelligenceapproachrevealedacomplexgenomicsignatureforcovid19severityinbrazilianpatients AT antoniomaurorezende humangenomepolymorphismsandcomputationalintelligenceapproachrevealedacomplexgenomicsignatureforcovid19severityinbrazilianpatients AT flaviorosendodasilvaoliveira humangenomepolymorphismsandcomputationalintelligenceapproachrevealedacomplexgenomicsignatureforcovid19severityinbrazilianpatients AT mariliadealbuquerquesena humangenomepolymorphismsandcomputationalintelligenceapproachrevealedacomplexgenomicsignatureforcovid19severityinbrazilianpatients AT clariceneuenschwanderlinsdemorais humangenomepolymorphismsandcomputationalintelligenceapproachrevealedacomplexgenomicsignatureforcovid19severityinbrazilianpatients AT cristianecampellobresanisalvi humangenomepolymorphismsandcomputationalintelligenceapproachrevealedacomplexgenomicsignatureforcovid19severityinbrazilianpatients AT luydsonrichardsonsilvavasconcelos humangenomepolymorphismsandcomputationalintelligenceapproachrevealedacomplexgenomicsignatureforcovid19severityinbrazilianpatients AT kennyadaniellecampelovalenca humangenomepolymorphismsandcomputationalintelligenceapproachrevealedacomplexgenomicsignatureforcovid19severityinbrazilianpatients AT carollinedearaujomariz humangenomepolymorphismsandcomputationalintelligenceapproachrevealedacomplexgenomicsignatureforcovid19severityinbrazilianpatients AT carlosbrito humangenomepolymorphismsandcomputationalintelligenceapproachrevealedacomplexgenomicsignatureforcovid19severityinbrazilianpatients AT claudioduartefonseca humangenomepolymorphismsandcomputationalintelligenceapproachrevealedacomplexgenomicsignatureforcovid19severityinbrazilianpatients AT cynthiabraga humangenomepolymorphismsandcomputationalintelligenceapproachrevealedacomplexgenomicsignatureforcovid19severityinbrazilianpatients AT christianrobsondesouzareis humangenomepolymorphismsandcomputationalintelligenceapproachrevealedacomplexgenomicsignatureforcovid19severityinbrazilianpatients AT ernestotorresdeazevedomarques humangenomepolymorphismsandcomputationalintelligenceapproachrevealedacomplexgenomicsignatureforcovid19severityinbrazilianpatients AT bartolomeuaciolisantos humangenomepolymorphismsandcomputationalintelligenceapproachrevealedacomplexgenomicsignatureforcovid19severityinbrazilianpatients |