The effect of migration and variation on populations of Escherichia coli adapting to complex fluctuating environments.

Migration, a critical evolutionary force, can have contrasting effects on adaptation. It can aid as well as impede adaptation. The effects of migration on microbial adaptation have been studied primarily in simple constant environments. Very little is known about the effects of migration on adaptation to complex, fluctuating environments. In our study, we subjected replicate populations of Escherichia coli, adapting to complex and unpredictably fluctuating environments to different proportions of clonal ancestral immigrants. Contrary to the results from simple/constant environments, the presence of clonal immigrants reduced all measured proxies of fitness. However, migration from a source population with a greater variance in fitness resulted in no change in fitness with respect to the no-migration control, except at the highest level of migration. Thus, the presence of variation in the immigrants could counter the adverse effects of migration in complex and unpredictably fluctuating environments. Our study demonstrates that the effects of migration are strongly dependent on the nature of the destination environment and the genetic makeup of immigrants. These results enhance our understanding of the influences of migrating populations, which could help better predict the consequences of migration.

An improved molecular inversion probe based targeted sequencing approach for low variant allele frequency.

Deep targeted sequencing technologies are still not widely used in clinical practice due to the complexity of the methods and their cost. The Molecular Inversion Probes (MIP) technology is cost effective and scalable in the number of targets, however, suffers from low overall performance especially in GC rich regions. In order to improve the MIP performance, we sequenced a large cohort of healthy individuals (n = 4417), with a panel of 616 MIPs, at high depth in duplicates. To improve the previous state-of-the-art statistical model for low variant allele frequency, we selected 4635 potentially positive variants and validated them using amplicon sequencing. Using machine learning prediction tools, we significantly improved precision of 10-56.25% (P < 0.0004) to detect variants with VAF > 0.005. We further developed biochemically modified MIP protocol and improved its turn-around-time to ∼4 h. Our new biochemistry significantly improved uniformity, GC-Rich regions coverage, and enabled 95% on target reads in a large MIP panel of 8349 genomic targets. Overall, we demonstrate an enhancement of the MIP targeted sequencing approach in both detection of low frequency variants and in other key parameters, paving its way to become an ultrafast cost-effective research and clinical diagnostic tool.