Plasmodium falciparum causes placental malaria, which results in adverse outcomes for mother and child. P. falciparum infected erythrocytes that express the parasite protein VAR2CSA on their surface can bind to placental chondroitin sulfate-A.  It has been hypothesized that naturally acquired antibodies towards VAR2CSA protect against placental infection, but it has proven difficult to use measures of antibody to identify individuals protected from disease. We used a systems serology approach to identify naturally acquired antibody features mid pregnancy that were associated with protection from placental malaria at delivery.  Machine learning techniques selected six out of 169 measured antibody features towards VAR2CSA that could predict (with 86% accuracy) whether a woman would subsequently have active placental malaria infection at delivery. Selected features were associated with inhibition of placental binding and/or opsonic phagocytosis of infected erythrocytes, and network analysis indicated that there are not one but multiple pathways to protection from placental malaria.

167 features towards chondroitin sulfate-A binding Plasmodium falciparum infected erythrocytes and individual domains of VAR2CSA recombinant protein were measured in 127 women at 14-26 gestation weeks.  Women were also grouped based on infection status at delivery.

After acquisition of antibody feature measurements the data was processed.  The right-skewness of the distribution of the features was reduced by log-transformation (log(x+1)). Four antibody features that had negative values were right-shifted to have their minimum at zero prior to log-transformation. Next, the distributions of the features were centered and scaled to have zero mean and unit standard deviation.   0.82% of observations were missing and were imputed, multivariate Imputations by Chained Equations (van Buuren and Groothuis-Oudshoorn, 2011) with predictive mean matching was used to impute any missing values. The imputation process was repeated five times and the median of the imputed values across the five generated imputed datasets was finally used for each missing value.

A detailed description of all variables is included in the ReadMe files.