CRED: A Deep Residual Network of Convolutional and Recurrent Units for Earthquake Signal Detection.

Earthquake signal detection is at the core of observational seismology. A good detection algorithm should be sensitive to small and weak events with a variety of waveform shapes, robust to background noise and non-earthquake signals, and efficient for processing large data volumes. Here, we introduce the Cnn-Rnn Earthquake Detector (CRED), a detector based on deep neural networks. CRED uses a combination of convolutional layers and bi-directional long-short-term memory units in a residual structure. It learns the time-frequency characteristics of the dominant phases in an earthquake signal from three component data recorded on individual stations. We train the network using 500,000 seismograms (250k associated with tectonic earthquakes and 250k identified as noise) recorded in Northern California. The robustness of the trained model with respect to the noise level and non-earthquake signals is shown by applying it to a set of semi-synthetic signals. We also apply the model to one month of continuous data recorded at Central Arkansas to demonstrate its efficiency, generalization, and sensitivity. Our model is able to detect more than 800 microearthquakes as small as -1.3 ML induced during hydraulic fracturing far away than the training region. We compare the performance of the model with the STA/LTA, template matching, and FAST algorithms. Our results indicate an efficient and reliable performance of CRED. This framework holds great promise for lowering the detection threshold while minimizing false positive detection rates.

Isolating single stranded DNA using a microfluidic dialysis device.

Isolating a particular strand of DNA from a double stranded DNA duplex is an important step in aptamer generation as well as many other biotechnology applications. Here we describe a microfluidic, flow-through, dialysis device for isolating single-stranded DNA (ssDNA) from double-stranded DNA (dsDNA). The device consists of two channels fabricated in polydimethylsiloxane (PDMS) separated by a track etched polycarbonate membrane (800 nm pore size). To isolate ssDNA, dual-biotin labelled dsDNA was immobilized onto streptavidin-coated polystyrene beads. Alkaline treatment was used to denature dsDNA, releasing the non-biotinylated ssDNA. In the flow-through dialysis device the liberated ssDNA was able to cross the membrane and was collected in an outlet channel. The complementary sequence bound to the bead was unable to cross the membrane and was directed to a waste channel. The effect of NaOH concentration and flow rate on purity and yield were compared. >95% ssDNA purity was achieved at 25 mM NaOH. However, lower flow rates were necessary to achieve ssDNA yields approaching the 50% theoretical maximum of the concurrent-flow device. Under optimized conditions the microfluidic isolation achieved even higher purity ssDNA than analogous manual procedures.

Size selective DNA transport through a nanoporous membrane in a PDMS microfluidic device.

A microfluidic counter current dialysis device for size based purification of DNA is described. The device consists of two polydimethylsiloxane (PDMS) channels separated by a track etched polycarbonate membrane with a 50 nm pore size. Recovery of fluorescein across the membrane was compared with 10 and 80 nucleotide (nt) ssDNA to characterize the device. Recovery of all three analytes improved with decreasing flow rate. Size selectivity was observed. Greater than 2-fold selectivity between 10 nt and 80 nt ssDNA was observed at linear velocities less than 3mm s(-1). Increasing the ionic strength of the buffer increased transport across the membrane. Recovery of 80 nt ssDNA increased over 4-fold by adding 30 mM NaCl to the buffer. The effect was size dependent as 10 nt showed a smaller increase while the recovery of fluorescein was largely unaffected by increasing the ionic strength of the buffer.