We interviewed 128 women regularly during pregnancy and the first postnatal year. Psychiatric interviews identified eight 'cases' of psychiatric disorder (6 per cent) in early pregnancy and twenty 'cases' (16 per cent) at six weeks after birth. Postnatal affective disorder, which accounted for 15 of these cases, was significantly associated with dissatisfaction with the marital relationship and also with previous psychiatric history. The implications of the term 'postnatal depression' are considered in terms of the course of the disorder in the 29 women (23 per cent) who had episodes of affective disorder at some time during pregnancy and the postnatal year. We found that the majority of episodes of affective disorder could be understood in terms of previous psychiatric history and/or reaction to life-events, including the stress of childbirth itself.

The binding and catalytic functions of proteins are generally mediated by a small number of functional residues held in place by the overall protein structure. Here, we describe deep learning approaches for scaffolding such functional sites without needing to prespecify the fold or secondary structure of the scaffold. The first approach, “constrained hallucination,” optimizes sequences such that their predicted structures contain the desired functional site. The second approach, “inpainting,” starts from the functional site and fills in additional sequence and structure to create a viable protein scaffold in a single forward pass through a specifically trained RoseTTAFold network. We use these two methods to design candidate immunogens, receptor traps, metalloproteins, enzymes, and protein-binding proteins and validate the designs using a combination of in silico and experimental tests.

There has been considerable recent progress in designing new proteins using deep-learning methods1–9. Despite this progress, a general deep-learning framework for protein design that enables solution of a wide range of design challenges, including de novo binder design and design of higher-order symmetric architectures, has yet to be described. Diffusion models10,11 have had considerable success in image and language generative modelling but limited success when applied to protein modelling, probably due to the complexity of protein backbone geometry and sequence–structure relationships. Here we show that by fine-tuning the RoseTTAFold structure prediction network on protein structure denoising tasks, we obtain a generative model of protein backbones that achieves outstanding performance on unconditional and topology-constrained protein monomer design, protein binder design, symmetric oligomer design, enzyme active site scaffolding and symmetric motif scaffolding for therapeutic and metal-binding protein design. We demonstrate the power and generality of the method, called RoseTTAFold diffusion (RFdiffusion), by experimentally characterizing the structures and functions of hundreds of designed symmetric assemblies, metal-binding proteins and protein binders. The accuracy of RFdiffusion is confirmed by the cryogenic electron microscopy structure of a designed binder in complex with influenza haemagglutinin that is nearly identical to the design model. In a manner analogous to networks that produce images from user-specified inputs, RFdiffusion enables the design of diverse functional proteins from simple molecular specifications.

Ordered two-dimensional arrays such as S-layers 1 , 2 and designed analogues 3 – 5 have intrigued bioengineers, 6 , 7 but with the exception of a single lattice formed with flexible linkers, 8 they are constituted from just one protein component. For modulating assembly dynamics and incorporating more complex functionality, materials composed of two components would have considerable advantages. 9 – 12 Here we describe a computational method to generate co-assembling binary layers by designing rigid interfaces between pairs of dihedral protein building-blocks, and use it to design a p6m lattice. The designed array components are soluble at mM concentrations, but when combined at nM concentrations, rapidly assemble into nearly crystalline micrometer-scale arrays nearly identical (based on TEM and SAXS) to the computational design model in vitro and in cells without the need for a two-dimensional support. Because the material is designed from the ground up, the components can be readily functionalized, and their symmetry reconfigured, enabling formation of ligand arrays with distinguishable surfaces which we demonstrate can drive extensive receptor clustering, downstream protein recruitment, and signaling. Using AFM on supported bilayers and quantitative microscopy on living cells, we show that arrays assembled on membranes have component stoichiometry and structure similar to arrays formed in vitro, and thus that our material can impose order onto fundamentally disordered substrates like cell membranes. In sharp contrast to previously characterized cell surface receptor binding assemblies such as antibodies and nanocages, which are rapidly endocytosed, we find that large arrays assembled at the cell surface suppress endocytosis in a tunable manner, with potential therapeutic relevance for extending receptor engagement and immune evasion. Our work paves the way towards a synthetic cell biology, where a new generation of multi-protein macroscale materials is designed to modulate cell responses and reshape synthetic and living systems.

No abstract

There has been considerable recent progress in designing new proteins using deep learning methods. Despite this progress, a general deep learning framework for protein design that enables solution of a wide range of design challenges, including de novo binder design and design of higher order symmetric architectures, has yet to be described. Diffusion models have had considerable success in image and language generative modeling but limited success when applied to protein modeling, likely due to the complexity of protein backbone geometry and sequence-structure relationships. Here we show that by fine tuning the RoseTTAFold structure prediction network on protein structure denoising tasks, we obtain a generative model of protein backbones that achieves outstanding performance on unconditional and topology-constrained protein monomer design, protein binder design, symmetric oligomer design, enzyme active site scaffolding, and symmetric motif scaffolding for therapeutic and metal-binding protein design. We demonstrate the power and generality of the method, called RoseTTAFold Diffusion (RFdiffusion), by experimentally characterizing the structures and functions of hundreds of new designs. In a manner analogous to networks which produce images from user-specified inputs, RFdiffusion enables the design of diverse, complex, functional proteins from simple molecular specifications.

Twelve judges, with no previous exposure to laryngectomees, rated the speaking proficiencies of 33 laryngectomees divided into the following groups: 1. esophageal speakers (n=12); 2. electrolarynx speakers (n=11); and 3. tracheoesophageal puncture speakers (n=10). In addition, the speech of ten normal subjects was rated. Specific speaking parameters examined included voice quality, pitch, loudness, intelligibility, rate of speaking, visual presentation during speech, extraneous speaking noise, and overall communicative effectiveness. Multiple discriminant analyses performed on the ratings made by each judge revealed significant differences in ratings for various speaking parameters in the four subject groups. Results generally support the stance that tracheoesophageal speech is perceived as superior to communication with either an electrolarynx or with traditional esophageal speech, even though it is not viewed as comparable to normal voice.

The Autism-Spectrum Quotient (AQ) has been developed to measure the degree to which an adult with normal intelligence has autistic traits. Although use of the AQ has resulted in a number of important findings, few studies have assessed whether scores predict cognitive aspects of ASD. This study assessed whether AQ scores predicted performance on an adapted block design. The test was adapted with a 'whole' and a 'segmented' task. High AQ scorers performed better than low scorers on the 'whole' task in the block design but performed equivalently on the 'segmented' task, as would be predicted in the autism spectrum. These findings add to the evidence showing construct validity for the AQ.