Total mutational load and clinical features as predictors of the metastatic status in lung adenocarcinoma and squamous cell carcinoma patients.

BACKGROUND: Recently, extensive cancer genomic studies have revealed mutational and clinical data of large cohorts of cancer patients. For example, the Pan-Lung Cancer 2016 dataset (part of The Cancer Genome Atlas project), summarises the mutational and clinical profiles of different subtypes of Lung Cancer (LC). Mutational and clinical signatures have been used independently for tumour typification and prediction of metastasis in LC patients. Is it then possible to achieve better typifications and predictions when combining both data streams? METHODS: In a cohort of 1144 Lung Adenocarcinoma (LUAD) and Lung Squamous Cell Carcinoma (LSCC) patients, we studied the number of missense mutations (hereafter, the Total Mutational Load TML) and distribution of clinical variables, for different classes of patients. Using the TML and different sets of clinical variables (tumour stage, age, sex, smoking status, and packs of cigarettes smoked per year), we built Random Forest classification models that calculate the likelihood of developing metastasis. RESULTS: We found that LC patients different in age, smoking status, and tumour type had significantly different mean TMLs. Although TML was an informative feature, its effect was secondary to the "tumour stage" feature. However, its contribution to the classification is not redundant with the latter; models trained using both TML and tumour stage performed better than models trained using only one of these variables. We found that models trained in the entire dataset (i.e., without using dimensionality reduction techniques) and without resampling achieved the highest performance, with an F1 score of 0.64 (95%CrI [0.62, 0.66]). CONCLUSIONS: Clinical variables and TML should be considered together when assessing the likelihood of LC patients progressing to metastatic states, as the information these encode is not redundant. Altogether, we provide new evidence of the need for comprehensive diagnostic tools for metastasis.

SCHC over LoRaWAN Efficiency: Evaluation and Experimental Performance of Packet Fragmentation.

Low Power Wide Area Networks (LPWAN) are expected to enable the massive connectivity of small and constrained devices to the Internet of Things. Due to the restricted nature of both end devices and network links, LPWAN technologies employ network stacks where there is no interoperable network layer as a general case; instead, application data are usually placed directly into technology-specific two-layer frames. Besides not being able to run standard IP-based protocols at the end device, the lack of an IP layer also causes LPWAN segments to operate in an isolated manner, requiring middleboxes to interface non-IP LPWAN technologies with the IP world. The IETF has standardized a compression and fragmentation scheme, called Static Context Header Compression and Fragmentation (SCHC), which can compress and fragment IPv6 and UDP headers for LPWAN in a way that enables IP-based communications on the constrained end device. This article presents a model to determine the channel occupation efficiency based on the transmission times of SCHC messages in the upstream channel of a LoRaWAN™ link using the ACK-on-Error mode of standard SCHC. The model is compared against experimental data obtained from the transmission of packets that are fragmented using a SCHC over LoRaWAN implementation. This modeling provides a relationship between the channel occupancy efficiency, the spreading factor of LoRa™, and the probability of an error of a SCHC message. The results show that the model correctly predicts the efficiency in channel occupation for all spreading factors. Furthermore, the SCHC ACK-on-Error mode implementation for the upstream channel has been made fully available for further use by the research community.

BEST: a Shiny/R web-based application to easily retrieve cross-related enzyme functional parameters and information from BRENDA.

MOTIVATION: BRENDA is the largest enzyme functional database, containing information of 84 000 experimentally characterized enzyme entries. This database is an invaluable resource for researchers in the biological field, which classifies enzyme-related information in categories that are very useful to obtain specific functional and protein engineering information for enzyme families. However, the BRENDA web interface, the most used by researchers with a non-informatic background, does not allow the user to cross-reference data from different categories or sub-categories in the database. Obtaining information in an easy and fast way, in a friendly web interface, without the necessity to have a deep informatics knowledge, will facilitate and improve research in the enzymology and protein engineering field. RESULTS: We developed the Brenda Easy Search Tool (BEST), an interactive Shiny/R application that enables querying the BRENDA database for complex cross-tabulated characteristics, and retrieving enzyme-related parameters and information readily and efficiently, which can be used for the study of enzyme function or as an input for other bioinformatics tools. AVAILABILITY AND IMPLEMENTATION: BEST and its tutorial are freely available from https://pesb2.cl/best/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.