Productivity Measurement through IMU-Based Detailed Activity Recognition Using Machine Learning: A Case Study of Masonry Work.

Although measuring worker productivity is crucial, the measurement of the productivity of each worker is challenging due to their dispersion across various construction jobsites. This paper presents a framework for measuring productivity based on an inertial measurement unit (IMU) and activity classification. Two deep learning algorithms and three sensor combinations were utilized to identify and analyze the feasibility of the framework in masonry work. Using the proposed method, worker activity classification could be performed with a maximum accuracy of 96.70% using the convolutional neural network model with multiple sensors, and a minimum accuracy of 72.11% using the long short-term memory (LSTM) model with a single sensor. Productivity could be measured with an accuracy of up to 96.47%. The main contributions of this study are the proposal of a method for classifying detailed activities and an exploration of the effect of the number of IMU sensors used in measuring worker productivity.

A CCR4-NOT Transcription Complex, Subunit 1, CNOT1, Variant Associated with Holoprosencephaly.

Holoprosencephaly is the incomplete separation of the forebrain during embryogenesis. Both genetic and environmental etiologies have been determined for holoprosencephaly; however, a genetic etiology is not found in most cases. In this report, we present two unrelated individuals with semilobar holoprosencephaly who have the identical de novo missense variant in the gene CCR4-NOT transcription complex, subunit 1 (CNOT1). The variant (c.1603C>T [p.Arg535Cys]) is predicted to be deleterious and is not present in public databases. CNOT1 has not been previously associated with holoprosencephaly or other brain malformations. In situ hybridization analyses of mouse embryos show that Cnot1 is expressed in the prosencephalic neural folds at gestational day 8.25 during the critical period for subsequent forebrain division. Combining human and mouse data, we show that CNOT1 is associated with incomplete forebrain division.

Rare hypomorphic human variation in the heptahelical domain of SMO contributes to holoprosencephaly phenotypes.

Holoprosencephaly (HPE) is the most common congenital anomaly affecting the forebrain and face in humans and occurs as frequently as 1:250 conceptions or 1:10,000 livebirths. Sonic Hedgehog signaling molecule is one of the best characterized HPE genes that plays crucial roles in numerous developmental processes including midline neural patterning and craniofacial development. The Frizzled class G-protein coupled receptor Smoothened (SMO), whose signaling activity is tightly regulated, is the sole obligate transducer of Hedgehog-related signals. However, except for previous reports of somatic oncogenic driver mutations in human cancers (or mosaic tumors in rare syndromes), any potential disease-related role of SMO genetic variation in humans is largely unknown. To our knowledge, ours is the first report of a human hypomorphic variant revealed by functional testing of seven distinct nonsynonymous SMO variants derived from HPE molecular and clinical data. Here we describe several zebrafish bioassays developed and guided by a systems biology analysis. This analysis strategy, and detection of hypomorphic variation in human SMO, demonstrates the necessity of integrating the genomic variant findings in HPE probands with other components of the Hedgehog gene regulatory network in overall medical interpretations.

Functional analysis of Sonic Hedgehog variants associated with holoprosencephaly in humans using a CRISPR/Cas9 zebrafish model.

Genetic variation in the highly conserved Sonic Hedgehog (SHH) gene is one of the most common genetic causes for the malformations of the brain and face in humans described as the holoprosencephaly clinical spectrum. However, only a minor fraction of known SHH variants have been experimentally proven to lead to abnormal function. Employing a phenotypic rescue assay with synthetic human messenger RNA variant constructs in shha-/- knockout zebrafish, we evaluated 104 clinically reported in-frame and missense SHH variants. Our data helped us to classify them into loss of function variants (31), hypomorphic variants (33), and nonpathogenic variants (40). We discuss the strengths and weaknesses of currently accepted predictors of variant deleteriousness and the American College of Medical Genetics and Genomics guidelines for variant interpretation in the context of this functional model; furthermore, we demonstrate the robustness of model systems such as zebrafish as a rapid method to resolve variants of uncertain significance.

Loss-of-function mutations in FGF8 can be independent risk factors for holoprosencephaly.

The utilization of next generation sequencing has been shown to accelerate gene discovery in human disease. However, our confidence in the correct disease-associations of rare variants continues to depend on functional analysis. Here, we employ a sensitive assay of human FGF8 variants in zebrafish to demonstrate that the spectrum of isoforms of FGF8 produced by alternative splicing can provide key insights into the genetic susceptibility to human malformations. In addition, we describe novel mutations in the FGF core structure that have both subtle and profound effects on ligand posttranslational processing and biological activity. Finally, we solve a case of apparent digenic inheritance of novel variants in SHH and FGF8, two genes known to functionally coregulate each other in the developing forebrain, as a simpler case of FGF8 diminished function.

Phenotype delineation of ZNF462 related syndrome.

Zinc finger protein 462 (ZNF462) is a relatively newly discovered vertebrate specific protein with known critical roles in embryonic development in animal models. Two case reports and a case series study have described the phenotype of 10 individuals with ZNF462 loss of function variants. Herein, we present 14 new individuals with loss of function variants to the previous studies to delineate the syndrome of loss of function in ZNF462. Collectively, these 24 individuals present with recurring phenotypes that define a multiple congenital anomaly syndrome. Most have some form of developmental delay (79%) and a minority has autism spectrum disorder (33%). Characteristic facial features include ptosis (83%), down slanting palpebral fissures (58%), exaggerated Cupid's bow/wide philtrum (54%), and arched eyebrows (50%). Metopic ridging or craniosynostosis was found in a third of study participants and feeding problems in half. Other phenotype characteristics include dysgenesis of the corpus callosum in 25% of individuals, hypotonia in half, and structural heart defects in 21%. Using facial analysis technology, a computer algorithm applying deep learning was able to accurately differentiate individuals with ZNF462 loss of function variants from individuals with Noonan syndrome and healthy controls. In summary, we describe a multiple congenital anomaly syndrome associated with haploinsufficiency of ZNF462 that has distinct clinical characteristics and facial features.

Common genetic causes of holoprosencephaly are limited to a small set of evolutionarily conserved driver genes of midline development coordinated by TGF-ß, hedgehog, and FGF signaling.

Here, we applied targeted capture to examine 153 genes representative of all the major vertebrate developmental pathways among 333 probands to rank their relative significance as causes for holoprosencephaly (HPE). We now show that comparisons of variant transmission versus nontransmission among 136 HPE Trios indicates some reported genes now lack confirmation, while novel genes are implicated. Furthermore, we demonstrate that variation of modest intrinsic effect can synergize with these driver mutations as gene modifiers.

Dominant-negative kinase domain mutations in FGFR1 can explain the clinical severity of Hartsfield syndrome.

Mutations in FGFR1 have recently been associated with Hartsfield syndrome, a clinically distinct syndromic form of holoprosencephaly (HPE) with ectrodactly, which frequently includes combinations of craniofacial, limb and brain abnormalities not typical for classical HPE. Unrelated clinical conditions generally without craniofacial or multi-system malformations include Kallmann syndrome and idiopathic hypogonadotropic hypogonadism. FGFR1 is a principal cause for these less severe diseases as well. Here we demonstrate that of the nine FGFR1 mutations recently detected in our screen of over 200 HPE probands by next generation sequencing, only five distinct mutations in the kinase domain behave as dominant-negative mutations in zebrafish over-expression assays. Three FGFR1 mutations seen in HPE probands behave identical to wild-type FGFR1 in rescue assays, including one apparent de novo variation. Interestingly, in one HPE family, a deleterious FGFR1 allele was transmitted from one parent and a loss-of-function allele in FGF8 from the other parent to both affected daughters. This family is one of the clearest examples to date of gene:gene synergistic interactions causing HPE in humans.

Embryonic mesoderm and endoderm induction requires the actions of non-embryonic Nodal-related ligands and Mxtx2.

Vertebrate mesoderm and endoderm formation requires signaling by Nodal-related ligands from the TGFß superfamily. The factors that initiate Nodal-related gene transcription are unknown in most species and the relative contributions of Nodal-related ligands from embryonic, extraembryonic and maternal sources remain uncertain. In zebrafish, signals from the yolk syncytial layer (YSL), an extraembryonic domain, are required for mesoderm and endoderm induction, and YSL expression of nodal-related 1 (ndr1) and ndr2 accounts for a portion of this activity. A variable requirement of maternally derived Ndr1 for dorsal and anterior axis formation has also been documented. Here we show that Mxtx2 directly activates expression of ndr2 via binding to its first intron and is required for ndr2 expression in the YSL. Mxtx2 is also required for the Nodal signaling-independent expression component of the no tail a (ntla) gene, which is required for posterior (tail) mesoderm formation. Therefore, Mxtx2 defines a new pathway upstream of Nodal signaling and posterior mesoderm formation. We further show that the co-disruption of extraembryonic Ndr2, extraembryonic Ndr1 and maternal Ndr1 eliminates endoderm and anterior (head and trunk) mesoderm, recapitulating the loss of Nodal signaling phenotype. Therefore, non-embryonic sources of Nodal-related ligands account for the complete spectrum of early Nodal signaling requirements. In summary, the induction of mesoderm and endoderm depends upon the combined actions of Mxtx2 and Nodal-related ligands from non-embryonic sources.

FGF-dependent left-right asymmetry patterning in zebrafish is mediated by Ier2 and Fibp1.

Establishment of left-right asymmetry in vertebrates requires nodal, Wnt-PCP and FGF signaling and involves ciliogenesis in a laterality organ. Effector genes through which FGF signaling affects laterality have not been described. We isolated the zebrafish ier2 and fibp1 genes as FGF target genes and show that their protein products interact. Knock down of these factors interferes with establishment of organ laterality and causes defective cilia formation in Kupffer's Vesicle, the zebrafish laterality organ. Cilia are also lost after suppression of FGF8, but can be rescued by injection of ier2 and fibp1 mRNA. We conclude that Ier2 and Fibp1 mediate FGF signaling in ciliogenesis in Kupffer's Vesicle and in the establishment of laterality in the zebrafish embryo.