Evaluation of text-to-gesture generation model using convolutional neural network.

Conversational gestures have a crucial role in realizing natural interactions with virtual agents and robots. Data-driven approaches, such as deep learning and machine learning, are promising in constructing the gesture generation model, which automatically provides the gesture motion for speech or spoken texts. This study experimentally analyzes a deep learning-based gesture generation model from spoken text using a convolutional neural network. The proposed model takes a sequence of spoken words as the input and outputs a sequence of 2D joint coordinates representing the conversational gesture motion. We prepare a dataset consisting of gesture motions and spoken texts by adding text information to an existing dataset and train the models using specific speaker's data. The quality of the generated gestures is compared with those from an existing speech-to-gesture generation model through a user perceptual study. The subjective evaluation shows that the model performance is comparable or superior to those by the existing speech-to-gesture generation model. In addition, we investigate the importance of data cleansing and loss function selection in the text-to-gesture generation model. We further examine the model transferability between speakers. The experimental results demonstrate successful model transferability of the proposed model. Finally, we show that the text-to-gesture generation model can produce good quality gestures even when using a transformer architecture.

Latent 3D Volume for Joint Depth Estimation and Semantic Segmentation from a Single Image.

This paper proposes a novel 3D representation, namely, a latent 3D volume, for joint depth estimation and semantic segmentation. Most previous studies encoded an input scene (typically given as a 2D image) into a set of feature vectors arranged over a 2D plane. However, considering the real world is three-dimensional, this 2D arrangement reduces one dimension and may limit the capacity of feature representation. In contrast, we examine the idea of arranging the feature vectors in 3D space rather than in a 2D plane. We refer to this 3D volumetric arrangement as a latent 3D volume. We will show that the latent 3D volume is beneficial to the tasks of depth estimation and semantic segmentation because these tasks require an understanding of the 3D structure of the scene. Our network first constructs an initial 3D volume using image features and then generates latent 3D volume by passing the initial 3D volume through several 3D convolutional layers. We apply depth regression and semantic segmentation by projecting the latent 3D volume onto a 2D plane. The evaluation results show that our method outperforms previous approaches on the NYU Depth v2 dataset.

Rare PDCD11 variations are not associated with risk of schizophrenia in Japan.

AIM: Rare gene variations are thought to confer substantial risk for schizophrenia. We performed a three-stage study to identify rare variations that have a strong impact on the risk of developing schizophrenia. METHODS: In the first stage, we prioritized rare missense variations using whole-exome sequencing (WES) data from three families, consisting of a proband, an affected sibling, and parents. In the second stage, we performed targeted resequencing of the PDCD11 coding region in 96 patients. In the third stage, we conducted an association study of rare PDCD11 variations with schizophrenia in a total of 1357 patients and 1394 controls. RESULTS: Via WES, we identified two rare missense PDCD11 variations, p.(Asp961Asn) and p.(Val1240Leu), shared by two affected siblings within families. Targeted resequencing of the PDCD11 coding region identified three rare non-synonymous variations: p.(Asp961Asn), p.(Phe1835del), and p.(Arg1837His). The case-control study demonstrated no significant associations between schizophrenia and four rare PDCD11 variations: p.(Asp961Asn), p.(Val1240Leu), p.(Phe1835del), and p.(Arg1837His). CONCLUSION: Our data do not support the role of rare PDCD11 variations in conferring substantial risk for schizophrenia in the Japanese population.

Rare UNC13B variations and risk of schizophrenia: Whole-exome sequencing in a multiplex family and follow-up resequencing and a case-control study.

Rare genomic variations inherited in multiplex schizophrenia families are suggested to play a role in the genetic etiology of the disease. To identify rare variations with large effects on the risk of developing schizophrenia, we performed whole-exome sequencing (WES) in two affected and one unaffected individual of a multiplex family with 10 affected individuals. We also performed follow-up resequencing of the unc-13 homolog B (Caenorhabditis elegans) (UNC13B) gene, a potential risk gene identified by WES, in the multiplex family and undertook a case-control study to investigate association between UNC13B and schizophrenia. UNC13B coding regions (39 exons) from 15 individuals of the multiplex family and 111 affected offspring for whom parental DNA samples were available were resequenced. Rare missense UNC13B variations identified by resequencing were further tested for association with schizophrenia in two independent case-control populations comprising a total of 1,753 patients and 1,602 controls. A rare missense variation (V1525M) in UNC13B was identified by WES in the multiplex family; this variation was present in five of six affected individuals, but not in eight unaffected individuals or one individual of unknown disease status. Resequencing UNC13B coding regions identified five rare missense variations (T103M, M813T, P1349T, I1362T, and V1525M). In the case-control study, there was no significant association between rare missense UNC13B variations and schizophrenia, although single-variant meta-analysis indicated that M813T was nominally associated with schizophrenia. These results do not support a contribution of rare missense UNC13B variations to the genetic etiology of schizophrenia in the Japanese population. © 2016 Wiley Periodicals, Inc.

Rare truncating variations and risk of schizophrenia: Whole-exome sequencing in three families with affected siblings and a three-stage follow-up study in a Japanese population.

Rare inherited variations in multiplex families with schizophrenia are suggested to play a role in the genetic etiology of schizophrenia. To further investigate the role of rare inherited variations, we performed whole-exome sequencing (WES) in three families, each with two affected siblings. We also performed a three-stage follow-up case-control study in a Japanese population with a total of 2617 patients and 2396 controls. WES identified 15 rare truncating variations that were variously present in the two affected siblings in each family. These variations did not necessarily segregate with schizophrenia within families, and they were different in each family. In the follow-up study, four variations (NWD1 W169X, LCORL R7fsX53, CAMK2B L497fsX497, and C9orf89 Q102X) had a higher mutant allele frequency in patients compared with controls, although these associations were not significant in the combined population, which comprised the first-, second- and third-stage populations. These results do not support a contribution of the rare truncating variations identified in the three families to the genetic etiology of schizophrenia.

Association analysis of putative cis-acting polymorphisms of interleukin-19 gene with schizophrenia.

BACKGROUND: Genome-wide association studies (GWAS) and gene expression analyses have revealed that single nucleotide polymorphisms (SNPs) associated with multifactorial diseases, such as schizophrenia, are significantly more likely to be associated with expression quantitative trait loci (eQTL). It was recently suggested that an immune system imbalance plays an important role in the pathogenesis of schizophrenia. Interleukin-19 is a novel cytokine that may play multiple roles in immune regulation and various diseases. METHOD: We selected eight tag SNPs in the eQTL of the IL-19 gene. Seven of the SNPs are putative cis-acting SNPs. Then, we conducted a case-control study using two independent samples. The first sample comprised 567 schizophrenia patients and 710 controls, and the second sample comprised 677 schizophrenia patients and 667 controls. RESULT: We identified the TGAA haplotype as being significantly associated with schizophrenia (p=0.0036 and corrected p=0.0264), although a combined analysis of the TGAA haplotype with the replication samples exhibited a nominally significant difference (p=0.022 and corrected p=0.235). CONCLUSIONS: These results suggest that the IL-19 gene might slightly contribute to the genetic risk of schizophrenia. Thus, further research on the association of eQTL SNPs with schizophrenia is warranted.

Interleukin 1 beta gene and risk of schizophrenia: detailed case-control and family-based studies and an updated meta-analysis.

OBJECTIVE: Interleukin-1 beta (IL-1ß) has been implicated in the pathophysiology of schizophrenia. To assess whether the IL1B gene confers increased susceptibility to schizophrenia, we conducted case-control and family-based studies and an updated meta-analysis. METHODS: We tested the association between IL1B and schizophrenia in 1229 case-control and 112 trio samples using 12 markers, including common tagging single nucleotide variations (SNVs) and a rare non-synonymous variation detected by resequencing the coding regions. We also performed a meta-analysis of rs16944 using a total of 8724 case-control and 201 trio samples from 16 independent populations. RESULTS: We found no significant associations between any of the 12 SNVs examined and schizophrenia in either case-control or trio samples. Moreover, our meta-analysis results showed no significant association between the common SNV, rs16944, and schizophrenia. CONCLUSIONS: The present study does not support a role for IL1B in schizophrenia susceptibility.

Resequencing and association analysis of MIR137 with schizophrenia in a Japanese population.

MicroRNA may play a role in the pathophysiology of schizophrenia. A recent meta-analysis of genome-wide association studies indicated a significant association between schizophrenia and a common intronic variation in MIR137HG (microRNA 137 host gene) encoding the primary microRNA-137. To explore additional risk variations for schizophrenia, we resequenced MIR137 and performed an association analysis in 1321 Japanese individuals. By resequencing, we detected four sequence variations in the 5' and 3' flanking regions. There were no significant associations between these variations and schizophrenia. Our resequencing and association analysis of MIR137 failed to find additional risk variations for schizophrenia.

A detailed association analysis between the tryptophan hydroxylase 2 (TPH2) gene and autism spectrum disorders in a Japanese population.

We conducted a detailed association analysis between the tryptophan hydroxylase 2 gene and autism spectrum disorders in a Japanese population using 19 markers, including tagging single nucleotide polymorphisms and a novel missense variation, p.R225Q, identified through exon resequencing. However, we failed to obtain supportive evidence for an association.

Failure to find an association between myosin heavy chain 9, non-muscle (MYH9) and schizophrenia: a three-stage case-control association study.

Several genome-wide linkage studies have suggested linkage between markers on the long arm of chromosome 22 and schizophrenia. It has also been reported that 22q11.2 deletions increase the risk of schizophrenia. Therefore, 22q is a candidate region for schizophrenia. To search for genetic susceptibility loci for schizophrenia on 22q, we conducted a three-stage case-control association study in Japanese individuals. In the first stage, we examined 13 microsatellite markers on 22q in 766 individuals (340 patients with schizophrenia and 426 control individuals) and found a potential association of AFM262VH5 (D22S283) with schizophrenia. In the second stage, we performed fine mapping of the myosin heavy chain 9, non-muscle (MYH9) gene, where AFM262VH5 is located, using 25 tagging single nucleotide polymorphisms (SNPs). We obtained potential associations between three SNPs in MYH9 and schizophrenia in 1193 individuals (595 patients and 598 controls), which included the individuals analyzed in the first stage. In the third stage, however, we could not replicate these associations in 4694 independent individuals (2288 patients and 2406 controls). Our results suggest that MYH9 does not confer increased susceptibility to schizophrenia in the Japanese population, although we could not exclude possible contributions of other genes on 22q to the pathogenesis of schizophrenia.

A case-control association analysis of CABIN1 with schizophrenia in a Japanese population.

Calcineurin (CN) is a calcium/calmodulin-dependent serine/threonine protein phosphatase and regulates neuronal structure, neurotransmission and activity-dependent gene expression. Several studies have indicated that CN signaling is likely to be involved in the pathogenesis of schizophrenia. The gene encoding CN-binding protein 1 (CABIN1) is located on 22q11.23, one of the common susceptibility loci for schizophrenia. Therefore, CABIN1 is a promising functional and positional candidate gene for schizophrenia. To assess whether CABIN1 is implicated in vulnerability to schizophrenia, we conducted a case-control association study between CABIN1 and schizophrenia. The results showed no evidence of an association between CABIN1 and schizophrenia using 11 tagging single nucleotide polymorphisms in 1193 Japanese subjects. Our results suggest that CABIN1 may not confer increased susceptibility for schizophrenia in the Japanese population.

The dopamine D3 receptor (DRD3) gene and risk of schizophrenia: case-control studies and an updated meta-analysis.

The dopamine D3 receptor (DRD3) has been suggested to be involved in the pathophysiology of schizophrenia. DRD3 has been tested for an association with schizophrenia, but with conflicting results. A recent meta-analysis suggested that the haplotype T-T-T-G for the SNPs rs7631540-rs1486012-rs2134655-rs963468 may confer protection against schizophrenia. However, almost all previous studies of the association between DRD3 and schizophrenia have been performed using a relatively small sample size and a limited number of markers. To assess whether DRD3 is implicated in vulnerability to schizophrenia, we conducted case-control association studies and performed an updated meta-analysis. In the first population (595 patients and 598 controls), we examined 16 genotyped single nucleotide polymorphisms (SNPs), including tagging SNPs selected from the HapMap database and SNPs detected through resequencing, as well as 58 imputed SNPs that are not directly genotyped. To confirm the results obtained, we genotyped the SNPs rs7631540-rs1486012-rs2134655-rs963468 in a second, independent population (2126 patients and 2228 controls). We also performed an updated meta-analysis of the haplotype, combining the results obtained in five populations, with a total sample size of 7551. No supportive evidence was obtained for an association between DRD3 and schizophrenia in our Japanese subjects. Our updated meta-analysis also failed to confirm the existence of a protective haplotype. To draw a definitive conclusion, further studies using larger samples and sufficient markers should be carried out in various ethnic populations.

A two-stage case-control association study of PADI2 with schizophrenia.

Peptidylarginine deiminases (PADIs), five isoforms of which have been identified, catalyze the conversion of arginine residues to citrulline residues in proteins. Recent studies have revealed that abnormal activation of PADI2, the gene for which is expressed throughout the nervous system, is likely to be related to the pathogenesis of neuropsychiatric diseases with neurodegenerative processes, such as Alzheimer's disease and multiple sclerosis. Such a progressive neurodegenerative process could be involved in the etiology and/or course of schizophrenia, and PADI2 may be a candidate gene for schizophrenia. To assess whether PADI2 has a role in vulnerability to schizophrenia, we conducted a two-stage case-control association study in Japanese individuals. In a screening population of 534 patients and 559 control individuals, we examined eight single-nucleotide polymorphisms (SNPs) including four haplotype tag SNPs and four coding SNPs in PADI2. There was a potential association of a synonymous SNP in exon 7 with schizophrenia. However, we could not replicate this association in a confirmatory population of 2126 patients and 2228 control individuals. The results of this study suggest that PADI2 does not contribute to genetic susceptibility to schizophrenia.

Preliminary genome-wide association study of bipolar disorder in the Japanese population.

Recent progress in genotyping technology and the development of public databases has enabled large-scale genome-wide association tests with diseases. We performed a two-stage genome-wide association study (GWAS) of bipolar disorder (BD) in Japanese cohorts. First we used Affymetrix 100K GeneChip arrays in the analysis of 107 cases with bipolar I disorder and 107 controls, and selected markers that were nominally significant (P < 0.01) in at least one of the three models (1,577 markers in total). In the follow-up stage, we analyzed these markers using an Illumina platform (1,526 markers; 51 markers were not designable for the platform) and an independent sample set, which consisted of 395 cases (bipolar I + II) and 409 controls. We also assessed the population stratification of current samples using principal components analysis. After the two-stage analysis, 89 markers remained nominally significant (allelic P < 0.05) with the same allele being consistently over-represented in both the first and the follow-up stages. However, none of these were significant after correction for multiple-testing by false discovery rates. Sample stratification was virtually negligible. Collectively, this is the first GWAS of BD in the Japanese population. But given the small sample size and the limited genomic coverage, these results should be taken as preliminary.