Linguistic aspects of speech synthesis.

The conversion of text to speech is seen as an analysis of the input text to obtain a common underlying linguistic description, followed by a synthesis of the output speech waveform from this fundamental specification. Hence, the comprehensive linguistic structure serving as the substrate for an utterance must be discovered by analysis from the text. The pronunciation of individual words in unrestricted text is determined by morphological analysis or letter-to-sound conversion, followed by specification of the word-level stress contour. In addition, many text character strings, such as titles, numbers, and acronyms, are abbreviations for normal words, which must be derived. To further refine these pronunciations and to discover the prosodic structure of the utterance, word part of speech must be computed, followed by a phrase-level parsing. From this structure the prosodic structure of the utterance can be determined, which is needed in order to specify the durational framework and fundamental frequency contour of the utterance. In discourse contexts, several factors such as the specification of new and old information, contrast, and pronominal reference can be used to further modify the prosodic specification. When the prosodic correlates have been computed and the segmental sequence is assembled, a complete input suitable for speech synthesis has been determined. Lastly, multilingual systems utilizing rule frameworks are mentioned, and future directions are characterized.

Frequency of HLA allele-specific peptide motifs in HIV-1 proteins correlates with the allele’s association with relative rates of disease progression after HIV-1 infection

An HLA allele-specific cytotoxic T lymphocyte response is thought to influence the rate of disease progression in HIV-1-infected individuals. In a prior study of 139 HIV-1-infected homosexual men, we identified HLA class I alleles and observed an association of specific alleles with different relative hazards for progression to AIDS. Seeking an explanation for this association, we searched HIV-1 protein sequences to determine the number of peptides matching motifs defined by combinations of specific amino acids reported to bind 16 class I alleles. Analyzing complete sequences of 12 clade B HIV isolates, we determined the number of allele motifs that were conserved (occurring in all 12 isolates) and nonconserved (occurring in only one isolate), as well as the average number of allele motifs per isolate. We found significant correlations with an allele’s association with disease progression for counts of conserved motifs in gag (R = 0.73; P = 0.002), pol (R = 0.58, P = 0.024), gp120 (R = 0.78, P = 0.00056), and total viral protein sequences (R = 0.67, P = 0.0058) and also for counts of nonconserved motifs in gag (R = 0.62, P = 0.013), pol (R = 0.74, P = 0.0017), gp41 (R = 0.52, P = 0.046), and total viral protein (R = 0.71, P = 0.0033). We also found significant correlations for the average number of motifs per isolate for gag, pol, gp120, and total viral protein. This study provides a plausible functional explanation for the observed association of different HLA alleles with variable rates of disease progression.

Voltage dependence of the pattern and frequency of discrete Ca2+ release events after brief repriming in frog skeletal muscle

Applying a brief repolarizing pre-pulse to a depolarized frog skeletal muscle fiber restores a small fraction of the transverse tubule membrane voltage sensors from the inactivated state. During a subsequent depolarizing test pulse we detected brief, highly localized elevations of myoplasmic Ca2+ concentration (Ca2+ “sparks”) initiated by restored voltage sensors in individual triads at all test pulse voltages. The latency histogram of these events gives the gating pattern of the sarcoplasmic reticulum (SR) calcium release channels controlled by the restored voltage sensors. Both event frequency and clustering of events near the start of the test pulse increase with test pulse depolarization. The macroscopic SR calcium release waveform, obtained from the spark latency histogram and the estimated open time of the channel or channels underlying a spark, exhibits an early peak and rapid marked decline during large depolarizations. For smaller depolarizations, the release waveform exhibits a smaller peak and a slower decline. However, the mean use time and mean amplitude of the individual sparks are quite similar at all test depolarizations and at all times during a given depolarization, indicating that the channel open times and conductances underlying sparks are essentially independent of voltage. Thus, the voltage dependence of SR Ca2+ release is due to changes in the frequency and pattern of occurrence of individual, voltage-independent, discrete release events.

Calcium electrogenesis in distal apical dendrites of layer 5 pyramidal cells at a critical frequency of back-propagating action potentials

Action potentials in juvenile and adult rat layer-5 neocortical pyramidal neurons can be initiated at both axonal and distal sites of the apical dendrite. However, little is known about the interaction between these two initiation sites. Here, we report that layer 5 pyramidal neurons are very sensitive to a critical frequency of back-propagating action potentials varying between 60 and 200 Hz in different neurons. Bursts of four to five back-propagating action potentials above the critical frequency elicited large regenerative potentials in the distal dendritic initiation zone. The critical frequency had a very narrow range (10–20 Hz), and the dendritic regenerative activity led to further depolarization at the soma. The dendritic frequency sensitivity was suppressed by blockers of voltage-gated calcium channels, and also by synaptically mediated inhibition. Calcium-fluorescence imaging revealed that the site of largest transient increase in intracellular calcium above the critical frequency was located 400–700 μm from the soma at the site for initiation of calcium action potentials. Thus, the distal dendritic initiation zone can interact with the axonal initiation zone, even when inputs to the neuron are restricted to regions close to the soma, if the output of the neuron exceeds a critical frequency.

Social transmission of reproductive behavior increases frequency of inherited disorders in a young-expanding population

The observation of high frequencies of certain inherited disorders in the population of Saguenay–Lac Saint Jean can be explained in terms of the variance and the correlation of effective family size (EFS) from one generation to the next. We have shown this effect by using the branching process approach with real demographic data. When variance of EFS is included in the model, despite its profound effect on mutant allele frequency, any mutant introduced in the population never reaches the known carrier frequencies (between 0.035 and 0.05). It is only when the EFS correlation between generations is introduced into the model that we can explain the rise of the mutant alleles. This correlation is described by a c parameter that reflects the dependency of children’s EFS on their parents’ EFS. The c parameter can be considered to reflect social transmission of demographic behavior. We show that such social transmission dramatically reduces the effective population size. This could explain particular distributions in allele frequencies and unusually high frequency of certain inherited disorders in some human populations.

Histone H1 Reduces the Frequency of Initiation in Xenopus Egg Extract by Limiting the Assembly of Prereplication Complexes on Sperm Chromatin

Somatic histone H1 reduces both the rate and extent of DNA replication in Xenopus egg extract. We show here that H1 inhibits replication directly by reducing the number of replication forks, but not the rate of fork progression, in Xenopus sperm nuclei. Density substitution experiments demonstrate that those forks that are active in H1 nuclei elongate to form large tracts of fully replicated DNA, indicating that inhibition is due to a reduction in the frequency of initiation and not the rate or extent of elongation. The observation that H1 dramatically reduces the number of replication foci in sperm nuclei supports this view. The establishment of replication competent DNA in egg extract requires the assembly of prereplication complexes (pre-RCs) on sperm chromatin. H1 reduces binding of the pre-RC proteins, XOrc2, XCdc6, and XMcm3, to chromatin. Replication competence can be restored in these nuclei, however, only under conditions that promote the loss of H1 from chromatin and licensing of the DNA. Thus, H1 inhibits replication in egg extract by preventing the assembly of pre-RCs on sperm chromatin, thereby reducing the frequency of initiation. These data raise the interesting possibility that H1 plays a role in regulating replication origin use during Xenopus development.

Effects of Genome Position and the DNA Damage Checkpoint on the Structure and Frequency of sod2 Gene Amplification in Fission Yeast

The Schizosaccharomyces pombe sod2 gene, located near the telomere on the long arm of chromosome I, encodes a Na+ (or Li+)/H+ antiporter. Amplification of sod2 has previously been shown to confer resistance to LiCl. We analyzed 20 independent LiCl-resistant strains and found that the only observed mechanism of resistance is amplification of sod2. The amplicons are linear, extrachromosomal elements either 225 or 180 kb long, containing both sod2 and telomere sequences. To determine whether proximity to a telomere is necessary for sod2 amplification, a strain was constructed in which the gene was moved to the middle of the same chromosomal arm. Selection of LiCl-resistant strains in this genetic background also yielded amplifications of sod2, but in this case the amplified DNA was exclusively chromosomal. Thus, proximity to a telomere is not a prerequisite for gene amplification in S. pombe but does affect the mechanism. Relative to wild-type cells, mutants with defects in the DNA damage aspect of the rad checkpoint control pathway had an increased frequency of sod2 amplification, whereas mutants defective in the S-phase completion checkpoint did not. Two models for generating the amplified DNA are presented.

Frequency of migrants and migratory activity are genetically correlated in a bird population: Evolutionary implications

Most migratory bird populations are composed of individuals that migrate and individuals that remain resident. While the role of ecological factors in maintaining this behavioral dimorphism has received much attention, the importance of genetic constraints on the evolution of avian migration has not yet been considered. Drawing on the recorded migratory activities of 775 blackcaps (Sylvia atricapilla) from a partially migratory population in southern France, we tested two alternative genetic models about the relationship between incidence and amount of migratory activity. The amount of migratory activity could be the continuous variable “underlying” the phenotypic expression of migratory urge, or, alternatively, the expression of both traits could be controlled by two separate genetic systems. The distributions of migratory activities in five different cohorts and the inheritance pattern derived from selective breeding experiments both indicate that incidence and amount of migratory activity are two aspects of one trait. Thus, all birds without measurable activity have activity levels at the low end of a continuous distribution, below the limit of expression or detection. The phenotypic dichotomy “migrant–nonmigrant” is caused by a threshold which may not be fixed but influenced both genetically and environmentally. This finding has profound implications for the evolution of migration: the transition from migratoriness to residency should not only be driven by selection favoring resident birds but also by selection for lower migratory activity. This potential for selection on two aspects, residency and migration distance, of the same trait may enable extremely rapid evolutionary changes to occur in migratory behavior.

Frequency of resistance to Bacillus thuringiensis in field populations of pink bollworm

Strategies for delaying pest resistance to genetically modified crops that produce Bacillus thuringiensis (Bt) toxins are based primarily on theoretical models. One key assumption of such models is that genes conferring resistance are rare. Previous estimates for lepidopteran pests targeted by Bt crops seem to meet this assumption. We report here that the estimated frequency of a recessive allele conferring resistance to Bt toxin Cry1Ac was 0.16 (95% confidence interval = 0.05–0.26) in strains of pink bollworm (Pectinophora gossypiella) derived from 10 Arizona cotton fields during 1997. Unexpectedly, the estimated resistance allele frequency did not increase from 1997 to 1999 and Bt cotton remained extremely effective against pink bollworm. These results demonstrate that the assumptions and predictions of resistance management models must be reexamined.

Inactivation of Saccharomyces cerevisiae OGG1 DNA repair gene leads to an increased frequency of mitochondrial mutants

The OGG1 gene encodes a highly conserved DNA glycosylase that repairs oxidized guanines in DNA. We have investigated the in vivo function of the Ogg1 protein in yeast mitochondria. We demonstrate that inactivation of ogg1 leads to at least a 2-fold increase in production of spontaneous mitochondrial mutants compared with wild-type. Using green fluorescent protein (GFP) we show that a GFP–Ogg1 fusion protein is transported to mitochondria. However, deletion of the first 11 amino acids from the N-terminus abolishes the transport of the GFP–Ogg1 fusion protein into the mitochondria. This analysis indicates that the N-terminus of Ogg1 contains the mitochondrial localization signal. We provide evidence that both yeast and human Ogg1 proteins protect the mitochondrial genome from spontaneous, as well as induced, oxidative damage. Genetic analyses revealed that the combined inactivation of OGG1 and OGG2 [encoding an isoform of the Ogg1 protein, also known as endonuclease three-like glycosylase I (Ntg1)] leads to suppression of spontaneously arising mutations in the mitochondrial genome when compared with the ogg1 single mutant or the wild-type. Together, these studies provide in vivo evidence for the repair of oxidative lesions in the mitochondrial genome by human and yeast Ogg1 proteins. Our study also identifies Ogg2 as a suppressor of oxidative mutagenesis in mitochondria.

Low frequency of E-cadherin alterations in familial breast cancer

In order to explore the possible role of E-cadherin in familial cancer, 19 familial breast cancer patients, whose tumours demonstrated loss of heterozygosity (LOH) at the E-cadherin locus, were screened for germline mutations. No pathogenic germline alterations were detected in these individuals. However, a somatic mutation was found (49-2A→C) in one of the tumours. This tumour showed a pattern of both ductal and lobular histology. Another 10 families with cases of breast, gastric and colon cancer were also screened for germline mutations, and no mutations were found. A missense mutation in exon 12 of E-cadherin (1774G→A; Ala592Thr) was previously found in one family with diffuse gastric cancer, and colon and breast cancer. An allelic association study was performed to determine whether the Ala592Thr alteration predisposes to breast cancer. In total, we studied 484 familial breast cancer patients, 614 sporadic breast cancer patients and 497 control individuals. The frequencies of this alteration were similar in these groups. However, a correlation between the Ala592Thr alteration and ductal comedo-type tumour was seen. These results, together with previously reported studies, indicate that germline mutations and, more commonly, somatic mutations in E-cadherin may have an influence on the behaviour of the tumours, rather than predispose to breast cancer.

High frequency of inactivating mutations in the neurofibromatosis type 2 gene (NF2) in primary malignant mesotheliomas.

Malignant mesotheliomas (MMs) are aggressive tumors that develop most frequently in the pleura of patients exposed to asbestos. In contrast to many other cancers, relatively few molecular alterations have been described in MMs. The most frequent numerical cytogenetic abnormality in MMs is loss of chromosome 22. The neurofibromatosis type 2 gene (NF2) is a tumor suppressor gene assigned to chromosome 22q which plays an important role in the development of familial and spontaneous tumors of neuroectodermal origin. Although MMs have a different histogenic derivation, the frequent abnormalities of chromosome 22 warranted an investigation of the NF2 gene in these tumors. Both cDNAs from 15 MM cell lines and genomic DNAs from 7 matched primary tumors were analyzed for mutations within the NF2 coding region. NF2 mutations predicting either interstitial in-frame deletions or truncation of the NF2-encoded protein (merlin) were detected in eight cell lines (53%), six of which were confirmed in primary tumor DNAs. In two samples that showed NF2 gene transcript alterations, no genomic DNA mutations were detected, suggesting that aberrant splicing may constitute an additional mechanism for merlin inactivation. These findings implicate NF2 in the oncogenesis of primary MMs and provide evidence that this gene can be involved in the development of tumors other than nervous system neoplasms characteristic of the NF2 disorder. In addition, unlike NF2-related tumors, MM derives from the mesoderm; malignancies of this origin have not previously been associated with frequent alterations of the NF2 gene.

Models of speech synthesis.

The term "speech synthesis" has been used for diverse technical approaches. In this paper, some of the approaches used to generate synthetic speech in a text-to-speech system are reviewed, and some of the basic motivations for choosing one method over another are discussed. It is important to keep in mind, however, that speech synthesis models are needed not just for speech generation but to help us understand how speech is created, or even how articulation can explain language structure. General issues such as the synthesis of different voices, accents, and multiple languages are discussed as special challenges facing the speech synthesis community.