Antibody Subclass and Glycosylation Shift Following Effective TB Treatment.

With an estimated 25% of the global population infected with Mycobacterium tuberculosis (Mtb), tuberculosis (TB) remains a leading cause of death by infectious diseases. Humoral immunity following TB treatment is largely uncharacterized, and antibody profiling could provide insights into disease resolution. Here we focused on the distinctive TB-specific serum antibody features in active TB disease (ATB) and compared them with latent TB infection (LTBI) or treated ATB (txATB). As expected, di-galactosylated glycan structures (lacking sialic acid) found on IgG-Fc differentiated LTBI from ATB, but also discriminated txATB from ATB. Moreover, TB-specific IgG4 emerged as a novel antibody feature that correlated with active disease, elevated in ATB, but significantly diminished after therapy. These findings highlight 2 novel TB-specific antibody changes that track with the resolution of TB and may provide key insights to guide TB therapy.

Publisher Correction: IFN-γ-independent immune markers of Mycobacterium tuberculosis exposure.

In the version of this article originally published, there was an error in the abstract. The word disease should not have been included in the sentence "These individuals were highly exposed to Mtb but tested negative disease by IFN-γ release assay and tuberculin skin test, 'resisting' development of classic LTBI". The sentence should have been "These individuals were highly exposed to Mtb but tested negative by IFN-γ release assay and tuberculin skin test, 'resisting' development of classic LTBI." The error has been corrected in the HTML and PDF versions of this article.

IFN-γ-independent immune markers of Mycobacterium tuberculosis exposure.

Exposure to Mycobacterium tuberculosis (Mtb) results in heterogeneous clinical outcomes including primary progressive tuberculosis and latent Mtb infection (LTBI). Mtb infection is identified using the tuberculin skin test and interferon-γ (IFN-γ) release assay IGRA, and a positive result may prompt chemoprophylaxis to prevent progression to tuberculosis. In the present study, we report on a cohort of Ugandan individuals who were household contacts of patients with TB. These individuals were highly exposed to Mtb but tested negative disease by IFN-γ release assay and tuberculin skin test, 'resisting' development of classic LTBI. We show that 'resisters' possess IgM, class-switched IgG antibody responses and non-IFN-γ T cell responses to the Mtb-specific proteins ESAT6 and CFP10, immunologic evidence of exposure to Mtb. Compared to subjects with classic LTBI, 'resisters' display enhanced antibody avidity and distinct Mtb-specific IgG Fc profiles. These data reveal a distinctive adaptive immune profile among Mtb-exposed subjects, supporting an expanded definition of the host response to Mtb exposure, with implications for public health and the design of clinical trials.

Hybrid computing using a neural network with dynamic external memory.

Artificial neural networks are remarkably adept at sensory processing, sequence learning and reinforcement learning, but are limited in their ability to represent variables and data structures and to store data over long timescales, owing to the lack of an external memory. Here we introduce a machine learning model called a differentiable neural computer (DNC), which consists of a neural network that can read from and write to an external memory matrix, analogous to the random-access memory in a conventional computer. Like a conventional computer, it can use its memory to represent and manipulate complex data structures, but, like a neural network, it can learn to do so from data. When trained with supervised learning, we demonstrate that a DNC can successfully answer synthetic questions designed to emulate reasoning and inference problems in natural language. We show that it can learn tasks such as finding the shortest path between specified points and inferring the missing links in randomly generated graphs, and then generalize these tasks to specific graphs such as transport networks and family trees. When trained with reinforcement learning, a DNC can complete a moving blocks puzzle in which changing goals are specified by sequences of symbols. Taken together, our results demonstrate that DNCs have the capacity to solve complex, structured tasks that are inaccessible to neural networks without external read-write memory.