Genomic epidemiology of SARS-CoV-2 in Spain

Conferencia presentada en el ESCMID Conference on Coronavirus Disease (ECCVID). Congreso Virtual, 23-25 de septiembre de 2020. La lista completa de personas e instituciones que forman parte de Seqcovid-Spain Consortium se puede ver en http://seqcovid.csic.es/collaborating-institutions/[Background] The rapid expansion of SARS-CoV-2 has represented a significant change in how scientists face a multidimensional problem. One of the most useful tools to find solutions to this problem is through the complete genome sequencing of viral samples. Hundreds of laboratories around the world have embarked in this task, sharing almost immediately their data. In Spain, we have established a multidisciplinary consortium including research laboratories and groups (the full list can be found at http://seqcovid.csic.es/collaborating-institutions/). Our initial goal is to obtain complete genome sequences of 20,000 viral samples from patients attending Spanish hospitals and health centers.[Methods] We have used remnants of diagnostic samples for Covid-19 to obtain almost complete viral genome sequences. We have essentially followed the ARTIC v3 protocol for Illumina sequencing, with some labs using MiniION or IonTorrent technologies.Sequences passing quality controls were uploaded to www.gisaid.org. Preliminary analyses were performed with NextStrain and Microrreact tools Additional analyses have been performed using IQTREE2 and iToL.[Results] We report results obtained until July 31, 2020, but an update summary is accessible at http://seqcovid.csic.es. We have received almost 6400 samples from which 3819 have been sequenced so far. Most sequenced samples were obtained in March (1674) and April (634). The epidemic in Spain is dominated by lineages B.1.5 (n=1250 ), A.2(907), B.1 (434), B.1.1 (336), and A.5 (273), representing 83.8% of the total (Figure 1). These samples conform at least 13 Spanish-dominated lineages (monophyletic groups of at least 20 sequences compared to aprox. 32000 complete genomes from GISAID with >50% sequences from Spain), most of which were already present in Spain when strict lockdown measures were enforced on March 14 (Figure 2). The D614 variant of the spike protein was found in 24% of the sequences, although its frequency has decreased significantly since April.[Conclusions] Similarly to other European countries, the Covid-19 epidemics in Spain is driven by a limited number of locally expanding lineages, which represent a fraction of the total number of independent introductions from other locations. The high frequency of lineages A.2 and A.5 point to a very early, mosty unnoticed entry of SARS-CoV-2 in Spain.Peer reviewe