Tracking SARS-CoV-2 introductions in Mozambique using pandemic-scale phylogenies: a retrospective observational study

BACKGROUND

From the start of the SARS-CoV-2 outbreak, global sequencing efforts have generated an unprecedented amount of genomic data. Nonetheless, unequal sampling between high-income and low-income countries hinders the implementation of genomic surveillance systems at the global and local level. Filling the knowledge gaps of genomic information and understanding pandemic dynamics in low-income countries is essential for public health decision making and to prepare for future pandemics. In this context, we aimed to discover the timing and origin of SARS-CoV-2 variant introductions in Mozambique, taking advantage of pandemic-scale phylogenies.

METHODS

We did a retrospective, observational study in southern Mozambique. Patients from Manhiça presenting with respiratory symptoms were recruited, and those enrolled in clinical trials were excluded. Data were included from three sources: (1) a prospective hospital-based surveillance study (MozCOVID), recruiting patients living in Manhiça, attending the Manhiça district hospital, and fulfilling the criteria of suspected COVID-19 case according to WHO; (2) symptomatic and asymptomatic individuals with SARS-CoV-2 infection recruited by the National Surveillance system; and (3) sequences from SARS-CoV-2-infected Mozambican cases deposited on the Global Initiative on Sharing Avian Influenza Data database. Positive samples amenable for sequencing were analysed. We used Ultrafast Sample placement on Existing tRees to understand the dynamics of beta and delta waves, using available genomic data. This tool can reconstruct a phylogeny with millions of sequences by efficient sample placement in a tree. We reconstructed a phylogeny (~7·6 million sequences) adding new and publicly available beta and delta sequences.

FINDINGS

A total of 5793 patients were recruited between Nov 1, 2020, and Aug 31, 2021. During this time, 133 328 COVID-19 cases were reported in Mozambique. 280 good quality new SARS-CoV-2 sequences were obtained after the inclusion criteria were applied and an additional 652 beta (B.1.351) and delta (B.1.617.2) public sequences were included from Mozambique. We evaluated 373 beta and 559 delta sequences. We identified 187 beta introductions (including 295 sequences), divided in 42 transmission groups and 145 unique introductions, mostly from South Africa, between August, 2020 and July, 2021. For delta, we identified 220 introductions (including 494 sequences), with 49 transmission groups and 171 unique introductions, mostly from the UK, India, and South Africa, between April and November, 2021.

INTERPRETATION

The timing and origin of introductions suggests that movement restrictions effectively avoided introductions from non-African countries, but not from surrounding countries. Our results raise questions about the imbalance between the consequences of restrictions and health benefits. This new understanding of pandemic dynamics in Mozambique can be used to inform public health interventions to control the spread of new variants.

FUNDING

European and Developing Countries Clinical Trials, European Research Council, Bill & Melinda Gates Foundation, and Agència de Gestió d'Ajuts Universitaris i de Recerca.

From 12 to 1 ECG lead: multiple cardiac condition detection mixing a hybrid machine learning approach with a one-vs-rest classification strategy

OBJECTIVE

Detecting different cardiac diseases using a single or reduced number of leads is still challenging. This work aims to provide and validate an automated method able to classify ECG recordings. Performance using complete 12-lead systems, reduced lead sets, and single-lead ECGs is evaluated and compared.

APPROACH

Seven different databases with 12-lead ECGs were provided during the PhysioNet/Computing in Cardiology Challenge 2021, where 88,253 annotated samples associated with none, one, or several cardiac conditions among 26 different classes were released for training, whereas 42,896 hidden samples were used for testing. After signal preprocessing, 81 features per ECG-lead were extracted, mainly based on heart rate variability, QRST patterns and spectral domain. Next, a One-vs-Rest classification approach made of independent binary classifiers for each cardiac condition was trained. This strategy allowed each ECG to be classified as belonging to none, one or several classes. For each class, a classification model among two binary Supervised Classifiers and one Hybrid Unsupervised-Supervised classification system was selected. Finally, we performed a 3-fold cross-validation to assess the system's performance.

MAIN RESULTS

Our classifiers received scores of 0.39, 0.38, 0.39, 0.38, and 0.37 for the 12, 6, 4, 3 and 2-lead versions of the hidden test set with the Challenge evaluation metric (CM). Also, we obtained a mean G-score of 0.80, 0.78, 0.79, 0.79, 0.77 and 0.74 for the 12, 6, 4, 3, 2 and 1-lead subsets with the public training set during our 3-fold cross-validation.

SIGNIFICANCE

We proposed and tested a machine learning approach focused on flexibility for identifying multiple cardiac conditions using one or more ECG leads. Our minimal-lead approach may be beneficial for novel portable or wearable ECG devices used as screening tools, as it can also detect multiple and concurrent cardiac conditions.

A Genomic Snapshot of the SARS-CoV-2 Pandemic in the Balearic Islands

Objective: To analyze the SARS-CoV-2 genomic epidemiology in the Balearic Islands, a unique setting in which the course of the pandemic has been influenced by a complex interplay between insularity, severe social restrictions and tourism travels.

Methods: Since the onset of the pandemic, more than 2,700 SARS-CoV-2 positive respiratory samples have been randomly selected and sequenced in the Balearic Islands. Genetic diversity of circulating variants was assessed by lineage assignment of consensus whole genome sequences with PANGOLIN and investigation of additional spike mutations.

Results: Consensus sequences were assigned to 46 different PANGO lineages and 75% of genomes were classified within a VOC, VUI, or VUM variant according to the WHO definitions. Highest genetic diversity was documented in the island of Majorca (42 different lineages detected). Globally, lineages B.1.1.7 and B.1.617.2/AY.X were identified as the 2 major lineages circulating in the Balearic Islands during the pandemic, distantly followed by lineages B.1.177/B.1.177.X. However, in Ibiza/Formentera lineage distribution was slightly different and lineage B.1.221 was the third most prevalent. Temporal distribution analysis showed that B.1 and B.1.5 lineages dominated the first epidemic wave, lineage B.1.177 dominated the second and third, and lineage B.1.617.2 the fourth. Of note, lineage B.1.1.7 became the most prevalent circulating lineage during first half of 2021; however, it was not associated with an increased in COVID-19 cases likely due to severe social restrictions and limited travels. Additional spike mutations were rarely documented with the exception of mutation S:Q613H which has been detected in several genomes (n = 25) since July 2021.

Conclusion: Virus evolution, mainly driven by the acquisition and selection of spike substitutions conferring biological advantages, social restrictions, and size population are apparently key factors for explaining the epidemic patterns registered in the Balearic Islands.

Complete Analysis of the Epidemiological Scenario around a SARS-CoV-2 Reinfection: Previous Infection Events and Subsequent Transmission

The first descriptions of reinfection by SARS-CoV-2 have been recently reported. However, these studies focus exclusively on the reinfected case, without considering the epidemiological context of the event. Our objectives were to perform a complete analysis of the sequential infections and community transmission events around a SARS-CoV-2 reinfection, including the infection events preceding it, the exposure, and subsequent transmissions. Our analysis was supported by host genetics, viral whole-genome sequencing, phylogenomic viral population analysis, and refined epidemiological data obtained from interviews with the involved subjects. The reinfection involved a 53-year-old woman with asthma (Case A), with a first COVID-19 episode in April 2020 and a much more severe second episode 4-1/2 months later, with SARS-CoV-2 seroconversion in August, that required hospital admission. An extended genomic analysis allowed us to demonstrate that the strain involved in Case A's reinfection was circulating in the epidemiological context of Case A and was also transmitted subsequently from Case A to her family context. The reinfection was also supported by a phylogenetic analysis, including 348 strains from Madrid, which revealed that the strain involved in the reinfection was circulating by the time Case A suffered the second episode, August-September 2020, but absent at the time range corresponding to Case A's first episode. IMPORTANCE We present the first complete analysis of the epidemiological scenario around a reinfection by SARS-CoV-2, more severe than the first episode, including three cases preceding the reinfection, the reinfected case per se, and the subsequent transmission to another seven cases.

Complete Analysis of the Epidemiological Scenario around a SARS-CoV-2 Reinfection: Previous Infection Events and Subsequent Transmission

The first descriptions of reinfection by SARS-CoV-2 have been recently reported. However, these studies focus exclusively on the reinfected case, without considering the epidemiological context of the event. Our objectives were to perform a complete analysis of the sequential infections and community transmission events around a SARS-CoV-2 reinfection, including the infection events preceding it, the exposure, and subsequent transmissions. Our analysis was supported by host genetics, viral whole-genome sequencing, phylogenomic viral population analysis, and refined epidemiological data obtained from interviews with the involved subjects. The reinfection involved a 53-year-old woman with asthma (Case A), with a first COVID-19 episode in April 2020 and a much more severe second episode 4-1/2 months later, with SARS-CoV-2 seroconversion in August, that required hospital admission. An extended genomic analysis allowed us to demonstrate that the strain involved in Case A's reinfection was circulating in the epidemiological context of Case A and was also transmitted subsequently from Case A to her family context. The reinfection was also supported by a phylogenetic analysis, including 348 strains from Madrid, which revealed that the strain involved in the reinfection was circulating by the time Case A suffered the second episode, August-September 2020, but absent at the time range corresponding to Case A's first episode. IMPORTANCE We present the first complete analysis of the epidemiological scenario around a reinfection by SARS-CoV-2, more severe than the first episode, including three cases preceding the reinfection, the reinfected case per se, and the subsequent transmission to another seven cases.

The first wave of the COVID-19 epidemic in Spain was associated with early introductions and fast spread of a dominating genetic variant

The coronavirus disease 2019 (COVID-19) pandemic has affected the world radically since 2020. Spain was one of the European countries with the highest incidence during the first wave. As a part of a consortium to monitor and study the evolution of the epidemic, we sequenced 2,170 samples, diagnosed mostly before lockdown measures. Here, we identified at least 500 introductions from multiple international sources and documented the early rise of two dominant Spanish epidemic clades (SECs), probably amplified by superspreading events. Both SECs were related closely to the initial Asian variants of SARS-CoV-2 and spread widely across Spain. We inferred a substantial reduction in the effective reproductive number of both SECs due to public-health interventions (R_e < 1), also reflected in the replacement of SECs by a new variant over the summer of 2020. In summary, we reveal a notable difference in the initial genetic makeup of SARS-CoV-2 in Spain compared with other European countries and show evidence to support the effectiveness of lockdown measures in controlling virus spread, even for the most successful genetic variants.

Atrial location optimization by electrical measures for Electrocardiographic Imaging

BACKGROUND

The Electrocardiographic Imaging (ECGI) technique, used to non-invasively reconstruct the epicardial electrical activity, requires an accurate model of the atria and torso anatomy. Here we evaluate a new automatic methodology able to locate the atrial anatomy within the torso based on an intrinsic electrical parameter of the ECGI solution.

METHODS

In 28 realistic simulations of the atrial electrical activity, we randomly displaced the atrial anatomy for ±2.5 cm and ±30° on each axis. An automatic optimization method based on the L-curve curvature was used to estimate the original position using exclusively non-invasive data.

RESULTS

The automatic optimization algorithm located the atrial anatomy with a deviation of 0.5 ± 0.5 cm in position and 16.0 ± 10.7° in orientation. With these approximate locations, the obtained electrophysiological maps reduced the average error in atrial rate measures from 1.1 ± 1.1 Hz to 0.5 ± 1.0 Hz and in the phase singularity position from 7.2 ± 4.0 cm to 1.6 ± 1.7 cm (p < 0.01).

CONCLUSIONS

This proposed automatic optimization may help to solve spatial inaccuracies provoked by cardiac motion or respiration, as well as to use ECGI on torso and atrial anatomies from different medical image systems.

A Mobile Health Application to Predict Postpartum Depression Based on Machine Learning

BACKGROUND

Postpartum depression (PPD) is a disorder that often goes undiagnosed. The development of a screening program requires considerable and careful effort, where evidence-based decisions have to be taken in order to obtain an effective test with a high level of sensitivity and an acceptable specificity that is quick to perform, easy to interpret, culturally sensitive, and cost-effective. The purpose of this article is twofold: first, to develop classification models for detecting the risk of PPD during the first week after childbirth, thus enabling early intervention; and second, to develop a mobile health (m-health) application (app) for the Android(®) (Google, Mountain View, CA) platform based on the model with best performance for both mothers who have just given birth and clinicians who want to monitor their patient's test.

MATERIALS AND METHODS

A set of predictive models for estimating the risk of PPD was trained using machine learning techniques and data about postpartum women collected from seven Spanish hospitals. An internal evaluation was carried out using a hold-out strategy. An easy flowchart and architecture for designing the graphical user interface of the m-health app was followed.

RESULTS

Naive Bayes showed the best balance between sensitivity and specificity as a predictive model for PPD during the first week after delivery. It was integrated into the clinical decision support system for Android mobile apps.

CONCLUSIONS

This approach can enable the early prediction and detection of PPD because it fulfills the conditions of an effective screening test with a high level of sensitivity and specificity that is quick to perform, easy to interpret, culturally sensitive, and cost-effective.