Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

General General

Synthesis of computer simulation and machine learning for achieving the best material properties of filled rubber.

In Scientific reports ; h5-index 158.0

Molecular dynamics (MD) simulation is used to analyze the mechanical properties of polymerized and nanoscale filled rubber. Unfortunately, the computation time for a simulation can require several months' computing power, because the interactions of thousands of filler particles must be calculated. To alleviate this problem, we introduce a surrogate convolutional neural network model to achieve faster and more accurate predictions. The major difficulty when employing machine-learning-based surrogate models is the shortage of training data, contributing to the huge simulation costs. To derive a highly accurate surrogate model using only a small amount of training data, we increase the number of training instances by dividing the large-scale simulation results into 3D images of middle-scale filler morphologies and corresponding regional stresses. The images include fringe regions to reflect the influence of the filler constituents outside the core regions. The resultant surrogate model provides higher prediction accuracy than that trained only by images of the entire region. Afterwards, we extract the fillers that dominate the mechanical properties using the surrogate model and we confirm their validity using MD.

Kojima Takashi, Washio Takashi, Hara Satoshi, Koishi Masataka

2020-Oct-22

General General

Prediction of rifampicin resistance beyond the RRDR using structure-based machine learning approaches.

In Scientific reports ; h5-index 158.0

Rifampicin resistance is a major therapeutic challenge, particularly in tuberculosis, leprosy, P. aeruginosa and S. aureus infections, where it develops via missense mutations in gene rpoB. Previously we have highlighted that these mutations reduce protein affinities within the RNA polymerase complex, subsequently reducing nucleic acid affinity. Here, we have used these insights to develop a computational rifampicin resistance predictor capable of identifying resistant mutations even outside the well-defined rifampicin resistance determining region (RRDR), using clinical M. tuberculosis sequencing information. Our tool successfully identified up to 90.9% of M. tuberculosis rpoB variants correctly, with sensitivity of 92.2%, specificity of 83.6% and MCC of 0.69, outperforming the current gold-standard GeneXpert-MTB/RIF. We show our model can be translated to other clinically relevant organisms: M. leprae, P. aeruginosa and S. aureus, despite weak sequence identity. Our method was implemented as an interactive tool, SUSPECT-RIF (StrUctural Susceptibility PrEdiCTion for RIFampicin), freely available at https://biosig.unimelb.edu.au/suspect_rif/ .

Portelli Stephanie, Myung Yoochan, Furnham Nicholas, Vedithi Sundeep Chaitanya, Pires Douglas E V, Ascher David B

2020-Oct-22

General General

Relationship between media multitasking and functional connectivity in the dorsal attention network.

In Scientific reports ; h5-index 158.0

With the development of digital technology, media multitasking behaviour, which is using two or more media simultaneously, has become more commonplace. There are two opposing hypotheses of media multitasking with regard to its impact on attention. One hypothesis claims that media multitasking can strengthen attention control, and the other claims heavy media multitaskers are less able to focus on relevant tasks in the presence of distractors. A total of 103 healthy subjects took part in this study. We measured the Media Multitasking Index (MMI) and subjects performed the continuous performance test. Resting state and oddball task functional MRI were conducted to analyse functional connectivity in the dorsal attention network, and the degree centrality (DC) was calculated using graph theory analysis. We found that the DCs in the dorsal attention network were higher during resting state than during the oddball task. Furthermore, the DCs during the task were positively correlated with the MMI. These results indicated that the DC reduction from resting state to the oddball task in high media multitaskers was attenuated compared with low media multitaskers. This study not only reveals more about the neurophysiology of media multitasking, but could also indicate brain biomarkers of media multitasking behaviour.

Kobayashi Kei, Oishi Naoya, Yoshimura Sayaka, Ueno Tsukasa, Miyagi Takashi, Murai Toshiya, Fujiwara Hironobu

2020-Oct-22

Public Health Public Health

Predicting the first smoking lapse during a quit attempt: A machine learning approach.

In Drug and alcohol dependence ; h5-index 64.0

BACKGROUND : Just-in-time adaptive interventions (JITAI) aim to prevent smoking lapse using tailored support delivered via mobile technology in the moments when it is most needed. Effective smoking cessation JITAI rely on the development of accurate decision rules that determine when someone is most likely to lapse. The primary goal of the present study was to identify the strongest predictors of first lapse among smokers undergoing a quit attempt.

METHODS : Smokers attending a clinic-based smoking cessation program (n = 74) were asked to complete ecological momentary assessments five times daily on study-provided smartphones for 4 weeks post-quit. A three-stage modeling process utilized Cox proportional hazards regression to examine time to lapse a function of 31 predictors. First, univariate models evaluated the relationship between each predictor and time to lapse. Second, the elastic net machine learning algorithm was used to select the best predictors. Third, backwards elimination further reduced the set of predictors to optimize parsimony.

RESULTS : Univariate models identified seven predictors significantly related to time to lapse. The elastic net algorithm retained five: perceived odds of smoking today, confidence in ability to avoid smoking, motivation to avoid smoking, urge to smoke, and cigarette availability. The reduced model demonstrated inadequate approximation to the non-penalized baseline model.

CONCLUSIONS : Accurate estimation of moments of high risk for smoking lapse remains an important goal in the development of JITAI. These results demonstrate the utility of exploratory data-driven approaches to variable selection. The results of this study can inform future JITAI by highlighting targets for intervention.

H├ębert Emily T, Suchting Robert, Ra Chaelin K, Alexander Adam C, Kendzor Darla E, Vidrine Damon J, Businelle Michael S

2020-Oct-11

Just-in-time adaptive intervention, Machine learning, Smartphones, Smoking cessation, mHealth

General General

Predicting enhancer-promoter interactions by deep learning and matching heuristic.

In Briefings in bioinformatics

Enhancer-promoter interactions (EPIs) play an important role in transcriptional regulation. Recently, machine learning-based methods have been widely used in the genome-scale identification of EPIs due to their promising predictive performance. In this paper, we propose a novel method, termed EPI-DLMH, for predicting EPIs with the use of DNA sequences only. EPI-DLMH consists of three major steps. First, a two-layer convolutional neural network is used to learn local features, and an bidirectional gated recurrent unit network is used to capture long-range dependencies on the sequences of promoters and enhancers. Second, an attention mechanism is used for focusing on relatively important features. Finally, a matching heuristic mechanism is introduced for the exploration of the interaction between enhancers and promoters. We use benchmark datasets in evaluating and comparing the proposed method with existing methods. Comparative results show that our model is superior to currently existing models in multiple cell lines. Specifically, we found that the matching heuristic mechanism introduced into the proposed model mainly contributes to the improvement of performance in terms of overall accuracy. Additionally, compared with existing models, our model is more efficient with regard to computational speed.

Min Xiaoping, Ye Congmin, Liu Xiangrong, Zeng Xiangxiang

2020-Oct-24

DNA sequence, deep learning, enhancer-promoter interactions, matching heuristic, pretraining

oncology Oncology

Stereotactic body radiation therapy for spinal metastases: a novel local control stratification by spinal region.

In Journal of neurosurgery. Spine

OBJECTIVE : This study evaluated a large cohort of patients treated with stereotactic body radiation therapy for spinal metastases and investigated predictive factors for local control, local progression-free survival (LPFS), overall survival, and pain response between the different spinal regions.

METHODS : The study was undertaken via retrospective review at a single institution. Patients with a tumor metastatic to the spine were included, while patients with benign tumors or primary spinal cord cancers were excluded. Statistical analysis involved univariate analysis, Cox proportional hazards analysis, the Kaplan-Meier method, and machine learning techniques (decision-tree analysis).

RESULTS : A total of 165 patients with 190 distinct lesions met all inclusion criteria for the study. Lesions were distributed throughout the cervical (19%), thoracic (43%), lumbar (19%), and sacral (18%) spines. The most common treatment regimen was 24 Gy in 3 fractions (44%). Via the Kaplan-Meier method, the 24-month local control was 80%. Sacral spine lesions demonstrated decreased local control (p = 0.01) and LPFS (p < 0.005) compared with those of the thoracolumbar spine. The cervical spine cases had improved local control (p < 0.005) and LPFS (p < 0.005) compared with the sacral spine and trended toward improvement relative to the thoracolumbar spine. The 36-month local control rates for cervical, thoracolumbar, and sacral tumors were 86%, 73%, and 44%, respectively. Comparably, the 36-month LPFS rates for cervical, thoracolumbar, and sacral tumors were 85%, 67%, and 35%, respectively. A planning target volume (PTV) > 50 cm3 was also predictive of local failure (p = 0.04). Fewer cervical spine cases had disease with PTV > 50 cm3 than the thoracolumbar (p = 5.87 × 10-8) and sacral (p = 3.9 × 10-3) cases. Using decision-tree analysis, the highest-fidelity models for predicting pain-free status and local failure demonstrated the first splits as being cervical and sacral location, respectively.

CONCLUSIONS : This study presents a novel risk stratification for local failure and LPFS by spinal region. Patients with metastases to the sacral spine may have decreased local control due to increased PTV, especially with a PTV of > 50 cm3. Multidisciplinary care should be emphasized in these patients, and both surgical intervention and radiotherapy should be strongly considered.

Kowalchuk Roman O, Waters Michael R, Richardson K Martin, Spencer Kelly, Larner James M, McAllister William H, Sheehan Jason P, Kersh Charles R

2020-Oct-23

BED = biologically effective dose, KPS = Karnofsky Performance Status, LPFS = local progression-free survival, PTV = planning target volume, SABR, SBRT, SBRT = stereotactic body radiation therapy, local control, metastasis, oncology, stereotactic ablative radiosurgery, stereotactic body radiation therapy