Extreme image transformations affect humans and machines differently.

Some recent artificial neural networks (ANNs) claim to model aspects of primate neural and human performance data. Their success in object recognition is, however, dependent on exploiting low-level features for solving visual tasks in a way that humans do not. As a result, out-of-distribution or adversarial input is often challenging for ANNs. Humans instead learn abstract patterns and are mostly unaffected by many extreme image distortions. We introduce a set of novel image transforms inspired by neurophysiological findings and evaluate humans and ANNs on an object recognition task. We show that machines perform better than humans for certain transforms and struggle to perform at par with humans on others that are easy for humans. We quantify the differences in accuracy for humans and machines and find a ranking of difficulty for our transforms for human data. We also suggest how certain characteristics of human visual processing can be adapted to improve the performance of ANNs for our difficult-for-machines transforms.

Machine Learning Heuristics on Gingivobuccal Cancer Gene Datasets Reveals Key Candidate Attributes for Prognosis.

Delayed cancer detection is one of the common causes of poor prognosis in the case of many cancers, including cancers of the oral cavity. Despite the improvement and development of new and efficient gene therapy treatments, very little has been carried out to algorithmically assess the impedance of these carcinomas. In this work, from attributes or NCBI's oral cancer datasets, viz. (i) name, (ii) gene(s), (iii) protein change, (iv) condition(s), clinical significance (last reviewed). We sought to train the number of instances emerging from them. Further, we attempt to annotate viable attributes in oral cancer gene datasets for the identification of gingivobuccal cancer (GBC). We further apply supervised and unsupervised machine learning methods to the gene datasets, revealing key candidate attributes for GBC prognosis. Our work highlights the importance of automated identification of key genes responsible for GBC that could perhaps be easily replicated in other forms of oral cancer detection.

Hypothetical Proteins as Predecessors of Long Non-coding RNAs.

Hypothetical Proteins [HP] are the transcripts predicted to be expressed in an organism, but no evidence of it exists in gene banks. On the other hand, long non-coding RNAs [lncRNAs] are the transcripts that might be present in the 5' UTR or intergenic regions of the genes whose lengths are above 200 bases. With the known unknown [KU] regions in the genomes rapidly existing in gene banks, there is a need to understand the role of open reading frames in the context of annotation. In this commentary, we emphasize that HPs could indeed be the predecessors of lncRNAs.

Electrophysiological Maturation of Cerebral Organoids Correlates with Dynamic Morphological and Cellular Development.

Cerebral organoids (COs) are rapidly accelerating the rate of translational neuroscience based on their potential to model complex features of the developing human brain. Several studies have examined the electrophysiological and neural network features of COs; however, no study has comprehensively investigated the developmental trajectory of electrophysiological properties in whole-brain COs and correlated these properties with developmentally linked morphological and cellular features. Here, we profiled the neuroelectrical activities of COs over the span of 5 months with a multi-electrode array platform and observed the emergence and maturation of several electrophysiologic properties, including rapid firing rates and network bursting events. To complement these analyses, we characterized the complex molecular and cellular development that gives rise to these mature neuroelectrical properties with immunohistochemical and single-cell transcriptomic analyses. This integrated approach highlights the value of COs as an emerging model system of human brain development and neurological disease.

A model to predict the function of hypothetical proteins through a nine-point classification scoring schema.

BACKGROUND: Hypothetical proteins [HP] are those that are predicted to be expressed in an organism, but no evidence of their existence is known. In the recent past, annotation and curation efforts have helped overcome the challenge in understanding their diverse functions. Techniques to decipher sequence-structure-function relationship, especially in terms of functional modelling of the HPs have been developed by researchers, but using the features as classifiers for HPs has not been attempted. With the rise in number of annotation strategies, next-generation sequencing methods have provided further understanding the functions of HPs. RESULTS: In our previous work, we developed a six-point classification scoring schema with annotation pertaining to protein family scores, orthology, protein interaction/association studies, bidirectional best BLAST hits, sorting signals, known databases and visualizers which were used to validate protein interactions. In this study, we introduced three more classifiers to our annotation system, viz. pseudogenes linked to HPs, homology modelling and non-coding RNAs associated to HPs. We discuss the challenges and performance of these classifiers using machine learning heuristics with an improved accuracy from Perceptron (81.08 to 97.67), Naive Bayes (54.05 to 96.67), Decision tree J48 (67.57 to 97.00), and SMO_npolyk (59.46 to 96.67). CONCLUSION: With the introduction of three new classification features, the performance of the nine-point classification scoring schema has an improved accuracy to functionally annotate the HPs.

---A web resource for nutrient use efficiency-related genes, quantitative trait loci and microRNAs in important cereals and model plants.

Cereals are key contributors to global food security. Genes involved in the uptake (transport), assimilation and utilization of macro- and micronutrients are responsible for the presence of these nutrients in grain and straw. Although many genomic databases for cereals are available, there is currently no cohesive web resource of manually curated nutrient use efficiency (NtUE)-related genes and quantitative trait loci (QTLs). In this study, we present a web-resource containing information on NtUE-related genes/QTLs and the corresponding available microRNAs for some of these genes in four major cereal crops (wheat ( Triticum aestivum), rice ( Oryza sativa), maize ( Zea mays), barley ( Hordeum vulgare)), two alien species related to wheat ( Triticum urartu and Aegilops tauschii), and two model species ( Brachypodium distachyon and Arabidopsis thaliana). Gene annotations integrated in the current web resource were manually curated from the existing databases and the available literature. The primary goal of developing this web resource is to provide descriptions of the NtUE-related genes and their functional annotation. MicroRNAs targeting some of the NtUE related genes and the QTLs for NtUE-related traits are also included. The genomic information embedded in the web resource should help users to search for the desired information.

HYPO: A Database of Human Hypothetical Proteins.

BACKGROUND: There are genes whose function remains obscure as they may not have similarities to known regions in the genome. Such known 'unknown' genes constituting the Open Reading Frames (ORF) that remain in the epigenome are termed as orphan genes and the proteins encoded by them but having no experimental evidence of translation are termed as 'Hypothetical Proteins' (HPs). OBJECTIVES: We have enhanced our former database of Hypothetical Proteins (HP) in human (HypoDB) with added annotation, application programming interfaces and descriptive features. The database hosts 1000+ manually curated records of the known 'unknown' regions in the human genome. The new updated version of HypoDB with functionalities (Blast, Match) is freely accessible at http://www.bioclues.org/hypo2. METHODS: The total collection of HPs were checked using experimentally validated sets (from Swiss-Prot) or non-experimentally validated set (TrEMBL) or the complete set (UniProtKB). The database was designed with java at the core backend, integrated with databases, viz. EMBL, PIR, HPRD and those including descriptors for structural databases, interaction and association databases. RESULTS: The HypoDB constituted Application Programming Interfaces (API) for implicitly searching resources linking them to other databases like NCBI Link-out in addition to multiple search capabilities along with advanced searches using integrated bio-tools, viz. Match and BLAST were incorporated. CONCLUSION: The HypoDB is perhaps the only open-source HP database with a range of tools for common bioinformatics retrievals and serves as a standby reference to researchers who are interested in finding candidate sequences for their potential experimental work.