Genomic and mobility data reveal mass population movement as a driver of SARS-CoV-2 dissemination and diversity in Bangladesh

Cowley, L. A.; Afrad, M. H.; Rahman, S. I. A.; Mahfuz-Al-Mamun, M.; Chin, T.; Mahmud, A. S.; Rahman, M. Z.; Billah, M. M.; Khan, M. H.; Sultana, S.; Khondaker, T.; Baker, S.; Banik, N.; Alam, A. N.; Mannoor, K.; Banu, S.; Chowdhury, A.; Flora, M. S.; Thomson, N.; Buckee, C.; Qadri, F.; Shirin, T.

Background New data streams are being used to track the pandemic of SARS-CoV-2, including genomic data which provides insights into patterns of importation and spatial spread of the virus, as well as population mobility data obtained from mobile phones. Here, we analyse the emergence and outbreak trajectory of SARS-CoV-2 in Bangladesh using these new data streams, and identify mass population movements as a key early event driving the ongoing epidemic. Methods We sequenced complete genomes of 67 SARS-CoV-2 samples (March-July 2020) and combined this dataset with 324 genomes from Bangladesh. For phylogenetic context, we also used 68,000 GISAID genomes collected globally. We paired this genomic data with population mobility information from Facebook and three mobile phone operators. Findings The majority (85%) of the Bangladeshi sequenced isolates fall into either pangolin lineage B.1.36 (8%), B.1.1 (19%) or B.1.1.25 (58%). Bayesian time-scaled phylogenetic analysis predicted SARS-COV-2 first appeared in mid-February, through international introductions. The first case was reported on March 8th. This pattern of repeated international introduction changed at the end of March when three discrete lineages expanded and spread clonally across Bangladesh. The shifting pattern of viral diversity across Bangladesh is reflected in the mobility data which shows the mass migration of people from cities to rural areas at the end of March, followed by frequent travel between Dhaka and the rest of the country during the following months. Interpretation In Bangladesh, population mobility out of Dhaka as well as frequent travel from urban hotspots to rural areas resulted in rapid country-wide dissemination of SARS-CoV-2. The strains in Bangladesh reflect the local expansion of global lineages introduced early from international travellers to and from major international travel hubs. Importantly, the Bangladeshi context is consistent with epidemiologic and phylogenetic findings globally. Bangladesh is one of the few countries in the world with a rich history of conducting mass vaccination campaigns under complex circumstances. Combining genomics and these new data streams should allow population movements to be modelled and anticipated rendering Bangladesh extremely well prepared to immunize citizens rapidly. Based on our genomics data and the country's successful immunization history, vaccines becoming available globally will be suitable for implementation in Bangladesh while ongoing genomic surveillance is conducted to monitor for new variants of the virus.