Scalable deep text comprehension for Cancer surveillance on high-performance computing.

BACKGROUND: Deep Learning (DL) has advanced the state-of-the-art capabilities in bioinformatics applications which has resulted in trends of increasingly sophisticated and computationally demanding models trained by larger and larger data sets. This vastly increased computational demand challenges the feasibility of conducting cutting-edge research. One solution is to distribute the vast computational workload across multiple computing cluster nodes with data parallelism algorithms. In this study, we used a High-Performance Computing environment and implemented the Downpour Stochastic Gradient Descent algorithm for data parallelism to train a Convolutional Neural Network (CNN) for the natural language processing task of information extraction from a massive dataset of cancer pathology reports. We evaluated the scalability improvements using data parallelism training and the Titan supercomputer at Oak Ridge Leadership Computing Facility. To evaluate scalability, we used different numbers of worker nodes and performed a set of experiments comparing the effects of different training batch sizes and optimizer functions. RESULTS: We found that Adadelta would consistently converge at a lower validation loss, though requiring over twice as many training epochs as the fastest converging optimizer, RMSProp. The Adam optimizer consistently achieved a close 2nd place minimum validation loss significantly faster; using a batch size of 16 and 32 allowed the network to converge in only 4.5 training epochs. CONCLUSIONS: We demonstrated that the networked training process is scalable across multiple compute nodes communicating with message passing interface while achieving higher classification accuracy compared to a traditional machine learning algorithm.

Non-functionalized carbon nanotube binding with hemoglobin.

Carbon nanotube has a high potential to be used as a biosensor and drug carrier. However, its binding behavior with proteins needs to be studied before the full potential of carbon nanotube in biological studies can be realized. Although many studies have been conducted to characterize the affinity of functionalized carbon nanotube to various types of proteins, our present study for the first time reported that hemoglobin can bind with non-functionalized carbon nanotube, and this binding can be identified by Raman spectrum. Further, this binding has not changed Raman luminescence with specific excitation and emission wavelengths. The immediate application of these findings is to use non-functionalized carbon nanotube as a biosensor to measure H(2)S in blood in which hemoglobin takes about 37% of the total blood volume. Other potential uses of non-functionalized carbon nanotube to bind selective groups of proteins are also hinted.