Robotic weed control has seen increased research of late with its potential for boosting productivity in agriculture. Majority of works focus on developing robotics for croplands, ignoring the weed management problems facing rangeland stock farmers. Perhaps the greatest obstacle to widespread uptake of robotic weed control is the robust classification of weed species in their natural environment. The unparalleled successes of deep learning make it an ideal candidate for recognising various weed species in the complex rangeland environment. This work contributes the first large, public, multiclass image dataset of weed species from the Australian rangelands; allowing for the development of robust classification methods to make robotic weed control viable. The DeepWeeds dataset consists of 17,509 labelled images of eight nationally significant weed species native to eight locations across northern Australia. This paper presents a baseline for classification performance on the dataset using the benchmark deep learning models, Inception-v3 and ResNet-50. These models achieved an average classification accuracy of 95.1% and 95.7%, respectively. We also demonstrate real time performance of the ResNet-50 architecture, with an average inference time of 53.4 ms per image. These strong results bode well for future field implementation of robotic weed control methods in the Australian rangelands.

Vici syndrome is a progressive neurodevelopmental multisystem disorder caused by mutations in the autophagy gene EPG5. Byrne et al. characterise the phenotype of 50 affected children, revealing callosal agenesis, cataracts, hypopigmentation, cardiomyopathy, immune dysfunction, developmental delay and microcephaly. Downregulation of epg5 in Drosophila results in autophagic abnormalities and progressive neurodegeneration.

The serotonin 2A receptor gene (HTR2A) harbors two functional single nucleotide polymorphisms (SNPs) that are frequent in populations of African and European descent; rs6311, which affects mRNA expression, and rs6314, which changes the amino acid sequence of the encoded protein and affects the signaling properties of the receptor. Multiple clinical associations support a role for these SNPs in cognitive and neuropsychiatric phenotypes, although studies in autism spectrum disorder (ASD) remain equivocal. Here, we tested transmission disequilibrium of rs6311 and rs6314 in a cohort of 158 ASD trios (simplex and multiplex), observing significant under-transmission of the minor “A” allele of rs6311 to offspring with ASD (permuted p=0.0004). Consistent with our previous findings in the dorsolateral prefrontal cortex of unaffected individuals, rs6311/A decreases expression of HTR2A mRNA with an extended 5′ untranslated region in the frontopolar cortex in brain samples from 54 ASD patients and controls. Interpreting the clinical results in the context of our mRNA expression analysis, we speculate that any risk associated with rs6311 is conferred by greater expression of the long 5′UTR mRNA isoform. The current study corroborates earlier associations between rs6311 and ASD in a family study, supporting the hypothesis that rs6311 plays a modulatory role in ASD risk.

Basan syndrome is an extremely rare ectodermal dysplasia with autosomal dominant inheritance and variable expressivity. The etiology of Basan syndrome remains unknown. To identify the Basan syndrome gene, we sequenced keratin 14 (KRT14) and SMARCAD1 in a previously unreported kindred with the disease. Sequencing of the coding regions and splice junctions of KRT14 and SMARCAD1 was performed using PCR-amplified genomic DNA isolated from blood or saliva and standard PCR protocols. In vitro functional studies were performed for a variant identified in SMARCAD1. While direct sequencing of KRT14 failed to reveal any likely pathogenic sequence alterations or splice site variants, a heterozygous splicing variant (c.378+3A>T) that segregated with the disease was identified in the skin-specific isoform of SMARCAD1. In vitro studies failed to demonstrate a splicing defect in SMARCAD1. We screened two candidate genes for Basan syndrome in a 3-generation pedigree. The skin-specific isoform of SMARCAD1 remains a good candidate for this disease.