Signal loss in blood oxygen level-dependent (BOLD) functional neuroimaging is common and can lead to misinterpretation of findings. Here, we reconstructed compromised fMRI signal using deep machine learning. We trained a model to learn principles governing BOLD activity in one dataset and reconstruct artificially compromised regions in an independent dataset, frame by frame. Intriguingly, BOLD time series extracted from reconstructed frames are correlated with the original time series, even though the frames do not independently carry any temporal information. Moreover, reconstructed functional connectivity maps exhibit good correspondence with the original connectivity maps, indicating that the model recovers functional relationships among brain regions. We replicated this result in two healthy datasets and in patients whose scans suffered signal loss due to intracortical electrodes. Critically, the reconstructions capture individual-specific information. Deep machine learning thus presents a unique opportunity to reconstruct compromised BOLD signal while capturing features of an individual's own functional brain organization.Oncogene-induced replication stress, for instance as a result of Cyclin E1 overexpression, causes genomic instability and has been linked to tumorigenesis. To survive high levels of replication stress, tumors depend on pathways to deal with these DNA lesions, which represent a therapeutically actionable vulnerability. We aimed to uncover the consequences of Cyclin E1 or Cdc25A overexpression on replication kinetics, mitotic progression, and the sensitivity to inhibitors of the WEE1 and ATR replication checkpoint kinases. We modeled oncogene-induced replication stress using inducible expression of Cyclin E1 or Cdc25A in non-transformed RPE-1 cells, either in a TP53 wild-type or TP53-mutant background. DNA fiber analysis showed Cyclin E1 or Cdc25A overexpression to slow replication speed. The resulting replication-derived DNA lesions were transmitted into mitosis causing chromosome segregation defects. Single cell sequencing revealed that replication stress and mitotic defects upon Cyclin E1 or Cdc25A overexpression resulted in genomic instability. ATR or WEE1 inhibition exacerbated the mitotic aberrancies induced by Cyclin E1 or Cdc25A overexpression, and caused cytotoxicity. Both these phenotypes were exacerbated upon p53 inactivation. Conversely, downregulation of Cyclin E1 rescued both replication kinetics, as well as sensitivity to ATR and WEE1 inhibitors. Taken together, Cyclin E1 or Cdc25A-induced replication stress leads to mitotic segregation defects and genomic instability. These mitotic defects are exacerbated by inhibition of ATR or WEE1 and therefore point to mitotic catastrophe as an underlying mechanism. Importantly, our data suggest that Cyclin E1 overexpression can be used to select patients for treatment with replication checkpoint inhibitors.Modifying material properties at the nanoscale is crucially important for devices in nano-electronics, nanophotonics and quantum information. Optically active defects in wide band gap materials, for instance, are critical constituents for the realisation of quantum technologies.  https://www.selleckchem.com/products/gsk2879552-2hcl.html  Here, we demonstrate the use of recoil implantation, a method exploiting momentum transfer from accelerated ions, for versatile and mask-free material doping. As a proof of concept, we direct-write arrays of optically active defects into diamond via momentum transfer from a Xe+ focused ion beam (FIB) to thin films of the group IV dopants pre-deposited onto a diamond surface. We further demonstrate the flexibility of the technique, by implanting rare earth ions into the core of a single mode fibre. We conclusively show that the presented technique yields ultra-shallow dopant profiles localised to the top few nanometres of the target surface, and use it to achieve sub-50 nm positional accuracy. The method is applicable to non-planar substrates with complex geometries, and it is suitable for applications such as electronic and magnetic doping of atomically-thin materials and engineering of near-surface states of semiconductor devices.Fluorine is a key element in the synthesis of molecules broadly used in medicine, agriculture and materials. Addition of fluorine to organic structures represents a unique strategy for tuning molecular properties, yet this atom is rarely found in Nature and approaches to integrate fluorometabolites into the biochemistry of living cells are scarce. In this work, synthetic gene circuits for organofluorine biosynthesis are implemented in the platform bacterium Pseudomonas putida. By harnessing fluoride-responsive riboswitches and the orthogonal T7 RNA polymerase, biochemical reactions needed for in vivo biofluorination are wired to the presence of fluoride (i.e. circumventing the need of feeding expensive additives). Biosynthesis of fluoronucleotides and fluorosugars in engineered P. putida is demonstrated with mineral fluoride both as only fluorine source (i.e. substrate of the pathway) and as inducer of the synthetic circuit. This approach expands the chemical landscape of cell factories by providing alternative biosynthetic strategies towards fluorinated building-blocks.Engineered RNA elements are programmable tools capable of detecting small molecules, proteins, and nucleic acids. Predicting the behavior of these synthetic biology components remains a challenge, a situation that could be addressed through enhanced pattern recognition from deep learning. Here, we investigate Deep Neural Networks (DNN) to predict toehold switch function as a canonical riboswitch model in synthetic biology. To facilitate DNN training, we synthesize and characterize in vivo a dataset of 91,534 toehold switches spanning 23 viral genomes and 906 human transcription factors. DNNs trained on nucleotide sequences outperform (R2 = 0.43-0.70) previous state-of-the-art thermodynamic and kinetic models (R2 = 0.04-0.15) and allow for human-understandable attention-visualizations (VIS4Map) to identify success and failure modes. This work shows that deep learning approaches can be used for functionality predictions and insight generation in RNA synthetic biology.