Homologous recombination repair gene mutations as a predictive biomarker for immunotherapy in patients with advanced melanoma.

Background: Nowadays, immunotherapy targeting immune checkpoint receptors is one of the cornerstones of systemic treatment in melanoma. Homologous recombination repair (HRR) is one of the DNA damage response (DDR) pathways, which has been proved to correlate with the efficacy of platinum-based chemotherapy, PARP inhibitor therapy, and immunotherapy in a variety of cancers. However, their predictive value of HRR remained unknown in patients with advanced melanoma.

Comparative Transcriptome Analysis Reveals Molecular Insights in Overwintering Monochamus alternatus (Coleoptera: Cerambycidae).

Monochamus alternatus, the dominant vector of Bursaphelenchus xylophilus (Aphelenchida: Aphelenchoididae), has caused immense damage to forest resources. In China, this vector was native to the southern regions but has spread northward recently. To adapt to more challenging environments in the northern winter, M.

Comparative Transcriptome Analysis of the Heat Stress Response in Hope (Coleoptera: Cerambycidae).

Temperature is a critical factor of insect population abundance and distribution. Hope (Coleoptera: Cerambycidae) is a significant concern since it is transmitted vector of the pinewood nematode posing enormous economic and environmental losses. This pest shows tolerance to heat stress, especially extremely high temperatures.

A Multi-scale Spatial-temporal Attention Model for Person Re-identification in Videos.

In this paper, we propose a novel deep neural network based attention model to learn the representative local regions from a video sequence for person re-identification. Specifically, we propose a multi-scale spatial-temporal attention (MSTA) model to measure the regions of each frame in different scales from the perspective of whole video sequence. Compared to traditional temporal attention models, MSTA focuses on exploiting the importance of local regions of each frame to the whole video representation in both spatial and temporal domains.

Feature Aggregation With Reinforcement Learning for Video-Based Person Re-Identification.

Video-based person re-identification (re-id) matches two tracks of persons from different cameras. Features are extracted from the images of a sequence and then aggregated as a track feature. Compared to existing works that aggregate frame features by simply averaging them or using temporal models such as recurrent neural networks, we propose an intelligent feature aggregate method based on reinforcement learning.