G-PLIP: Knowledge graph neural network for structure-free protein-ligand bioactivity prediction.

Protein-ligand interactions (PLIs) determine the efficacy and safety profiles of small molecule drugs. Existing methods rely on either structural information or resource-intensive computations to predict PLI, casting doubt on whether it is possible to perform structure-free PLI predictions at low computational cost. Here we show that a light-weight graph neural network (GNN), trained with quantitative PLIs of a small number of proteins and ligands, is able to predict the strength of unseen PLIs. The model has no direct access to structural information about the protein-ligand complexes. Instead, the predictive power is provided by encoding the entire chemical and proteomic space in a single heterogeneous graph, encapsulating primary protein sequence, gene expression, the protein-protein interaction network, and structural similarities between ligands. This novel approach performs competitively with, or better than, structure-aware models. Our results suggest that existing PLI prediction methods may be improved by incorporating representation learning techniques that embed biological and chemical knowledge.

Geometric deep learning-guided Suzuki reaction conditions assessment for applications in medicinal chemistry.

Suzuki cross-coupling reactions are considered a valuable tool for constructing carbon-carbon bonds in small molecule drug discovery. However, the synthesis of chemical matter often represents a time-consuming and labour-intensive bottleneck. We demonstrate how machine learning methods trained on high-throughput experimentation (HTE) data can be leveraged to enable fast reaction condition selection for novel coupling partners. We show that the trained models support chemists in determining suitable catalyst-solvent-base combinations for individual transformations including an evaluation of the need for HTE screening. We introduce an algorithm for designing 96-well plates optimized towards reaction yields and discuss the model performance of zero- and few-shot machine learning. The best-performing machine learning model achieved a three-category classification accuracy of 76.3% (±0.2%) and an F 1-score for a binary classification of 79.1% (±0.9%). Validation on eight reactions revealed a receiver operating characteristic (ROC) curve (AUC) value of 0.82 (±0.07) for few-shot machine learning. On the other hand, zero-shot machine learning models achieved a mean ROC-AUC value of 0.63 (±0.16). This study positively advocates the application of few-shot machine learning-guided reaction condition selection for HTE campaigns in medicinal chemistry and highlights practical applications as well as challenges associated with zero-shot machine learning.

Small-Molecule-Directed Endogenous Regeneration of Visual Function in a Mammalian Retinal Degeneration Model.

Degenerative retinal diseases associated with photoreceptor loss are a leading cause of visual impairment worldwide, with limited treatment options. Phenotypic profiling coupled with medicinal chemistry were used to develop a small molecule with proliferative effects on retinal stem/progenitor cells, as assessed in vitro in a neurosphere assay and in vivo by measuring Msx1-positive ciliary body cell proliferation. The compound was identified as having kinase inhibitory activity and was subjected to cellular pathway analysis in non-retinal human primary cell systems. When tested in a disease-relevant murine model of adult retinal degeneration (MNU-induced retinal degeneration), we observed that four repeat intravitreal injections of the compound improved the thickness of the outer nuclear layer along with the regeneration of the visual function, as measured with ERG, visual acuity, and contrast sensitivity tests. This serves as a proof of concept for the use of a small molecule to promote endogenous regeneration in the eye.

Enabling late-stage drug diversification by high-throughput experimentation with geometric deep learning.

Late-stage functionalization is an economical approach to optimize the properties of drug candidates. However, the chemical complexity of drug molecules often makes late-stage diversification challenging. To address this problem, a late-stage functionalization platform based on geometric deep learning and high-throughput reaction screening was developed. Considering borylation as a critical step in late-stage functionalization, the computational model predicted reaction yields for diverse reaction conditions with a mean absolute error margin of 4-5%, while the reactivity of novel reactions with known and unknown substrates was classified with a balanced accuracy of 92% and 67%, respectively. The regioselectivity of the major products was accurately captured with a classifier F-score of 67%. When applied to 23 diverse commercial drug molecules, the platform successfully identified numerous opportunities for structural diversification. The influence of steric and electronic information on model performance was quantified, and a comprehensive simple user-friendly reaction format was introduced that proved to be a key enabler for seamlessly integrating deep learning and high-throughput experimentation for late-stage functionalization.

Diamond preservation in the lithospheric mantle recorded by olivine in kimberlites.

The diamond potential of kimberlites is difficult to assess due to several mantle and magmatic processes affecting diamond content. Traditionally, initial evaluations are based on the compositions of mantle-derived minerals (garnet, chromite, clinopyroxene), which allow an assessment of pressure-temperature conditions and lithologies suitable for diamond formation. Here we explore a complementary approach that considers the conditions of diamonds destruction by interaction with melts/fluids (metasomatism). We test the hypothesis that carbonate-rich metasomatism related to kimberlite melt infiltration into the deep lithosphere is detrimental to diamond preservation. Our results show that high diamond grades in kimberlites worldwide are exclusively associated with high-Mg/Fe olivine, which corresponds to mantle lithosphere minimally affected by kimberlite-related metasomatism. Diamond dissolution in strongly metasomatised lithosphere containing low-Mg/Fe olivine provides a causal link to the empirical associations between low diamond grades, abundant Ti-Zr-rich garnets and kimberlites with high Ti and low Mg contents. This finding show-cases olivine geochemistry as a viable tool in diamond exploration.

Identifying opportunities for late-stage C-H alkylation with high-throughput experimentation and in silico reaction screening.

Enhancing the properties of advanced drug candidates is aided by the direct incorporation of specific chemical groups, avoiding the need to construct the entire compound from the ground up. Nevertheless, their chemical intricacy often poses challenges in predicting reactivity for C-H activation reactions and planning their synthesis. We adopted a reaction screening approach that combines high-throughput experimentation (HTE) at a nanomolar scale with computational graph neural networks (GNNs). This approach aims to identify suitable substrates for late-stage C-H alkylation using Minisci-type chemistry. GNNs were trained using experimentally generated reactions derived from in-house HTE and literature data. These trained models were then used to predict, in a forward-looking manner, the coupling of 3180 advanced heterocyclic building blocks with a diverse set of sp3-rich carboxylic acids. This predictive approach aimed to explore the substrate landscape for Minisci-type alkylations. Promising candidates were chosen, their production was scaled up, and they were subsequently isolated and characterized. This process led to the creation of 30 novel, functionally modified molecules that hold potential for further refinement. These results positively advocate the application of HTE-based machine learning to virtual reaction screening.

Sexual health behavior, health status, and knowledge among queer womxn and trans men in Kenya: An online cross-sectional study.

INTRODUCTION: Little research has been conducted on the sexual health needs and risk behaviors of queer womxn and trans men, making it difficult to identify specific health needs and disparities. This is especially the case in the Global South, where their needs are poorly understood. This study presents findings on demographics, sources of information, sexual (risk) behaviors, and substance use in Kenyan queer womxn and trans men. METHODS: An online survey among 335 Kenyan queer womxn and trans men was used to collect data on sexual health, risk behavior, health information sources, and substance use. The participants needed to have had at least one self-identified female sexual partner. RESULTS: The sample presented young, highly-educated queer womxn and trans men. A high incidence of childhood sexual trauma found was found. Risk behaviors included sexual activities with partners of multiple genders, violence, and low use of barrier methods. One in three participants had been treated for an STD in the previous year. The incidences of smoking and drinking were high, and a quarter of participants indicated having taken drugs at least once a month or more. The internet was either the first or second most important source of sexual health information for 44.1% of the participants, followed by schools (30.9%). DISCUSSION AND CONCLUSION: Our findings indicate that queer womxn and trans men are at risk of negative sexual health outcomes due to a lack of appropriate information, risk behavior, substance use, and low uptake of sexual health services. Kenya's Penal Code still criminalizes consensual same-sex activities and may play a role in perpetuating barriers that prohibit them from making healthier choices. Developing tailored programming and policies require local, national, and global stakeholders to engage with the inclusion of queer womxn and trans men's sexual health needs within strategic planning and healthcare delivery.

Using online spaces to recruit Kenyan queer womxn and trans men in restrictive offline settings.

BACKGROUND: Understanding and addressing healthcare and service delivery inequalities is essential to increase equity and overcome health disparities and service access discrimination. While tremendous progress has been made towards the inclusion of sexual and gender minorities in health and other research, gaps still exist. Innovative methods are needed to close these. This case study describes and reflects on using online-based data collection to ascertain sexual health decision-making and health service utilisation among Kenyan queer womxn and trans men. METHODS: Case study The study used a mixed-methods approach in two phases with triangulated quantitative and qualitative elements. Both elements used web-based technology to gather data. RESULTS: Using online spaces to recruit and collect data from queer womxn and trans men exceeded expectations. A total of 360 queer womxn and trans men responded to the digitally distributed survey, and 33 people, queer womxn and trans men, as well as key informants, participated in the interviews, which were primarily conducted on Zoom and Skype. The case study analyses the risks and benefits of this approach and concludes that online sampling approaches can mitigate risks and enable effective and safe sampling of a marginalised group in a restrictive legal setting: Kenyan queer womxn and trans men. CONCLUSION: Using online spaces when researching marginalised populations could effectively overcome risks around stigma, discrimination and violence. It could be an effective way to understand these populations' healthcare needs better. Factors contributing to success included building trusting relationships with key members of the community, strategic and opportune timing, a nuanced understanding of the mobile landscape, and carefully chosen safety and security measures. However, it should be noted that conducting research online could increase the risk of further marginalising and excluding those without access to web-based technology.

Experience of and factors associated with violence against sexual and gender minorities in nine African countries: a cross-sectional study.

OBJECTIVE: The objective of this research was to assess physical and sexual violence experienced by sexual and gender minorities in nine African countries, and to examine factors associated with violence. METHODS: We conducted an exploratory multi-country cross-sectional study among self-identifying sexual and gender minorities, using a survey tool available in paper and online. Participants were sampled through venue-based and web-based convenience sampling. We analysed data using descriptive statistics and logistic regression, with Stata15. FINDINGS: Of 3798 participants, 23% were gender minorities, 20% were living with HIV, and 18% had been coerced into marriage. Fifty-six per cent of all participants had experienced physical or sexual violence in their lifetime, and 29% in the past year. Gender minorities had experienced significantly higher levels of violence compared to cisgender (sexual minority) participants. The variable most strongly associated with having experienced violence was being coerced into marriage (AOR, 3.02), followed by people living nearby knowing about one's sexual orientation and/or gender identity (AOR, 1.90) and living with HIV (AOR, 1.47). CONCLUSION: Sexual and gender minorities in Eastern and Southern Africa experience high levels of violence. Sexual orientation and gender identity need to be recognised as risk factors for violence in national and regional law and policy frameworks. States should follow the African Commission Resolution 275 and provide protection against violence based on real or perceived sexual orientation or gender identity.

Morphing of Amphipathic Helices to Explore the Activity and Selectivity of Membranolytic Antimicrobial Peptides.

Naturally occurring membranolytic antimicrobial peptides (AMPs) are rarely cell-type selective and highly potent at the same time. Template-based peptide design can be used to generate AMPs with improved properties de novo. Following this approach, 18 linear peptides were obtained by computationally morphing the natural AMP Aurein 2.2d2 GLFDIVKKVVGALG into the synthetic model AMP KLLKLLKKLLKLLK. Eleven of the 18 chimeric designs inhibited the growth of Staphylococcus aureus, and six peptides were tested and found to be active against one resistant pathogenic strain or more. One of the peptides was broadly active against bacterial and fungal pathogens without exhibiting toxicity to certain human cell lines. Solution nuclear magnetic resonance and molecular dynamics simulation suggested an oblique-oriented membrane insertion mechanism of this helical de novo peptide. Temperature-resolved circular dichroism spectroscopy pointed to conformational flexibility as an essential feature of cell-type selective AMPs.

Debate: Why should gender-affirming health care be included in health science curricula?

BACKGROUND: Every person who seeks health care should be affirmed, respected, understood, and not judged. However, trans and gender diverse people have experienced significant marginalization and discrimination in health care settings. Health professionals are generally not adequately prepared by current curricula to provide appropriate healthcare to trans and gender diverse people. This strongly implies that health care students would benefit from curricula which facilitate learning about gender-affirming health care. MAIN BODY: Trans and gender diverse people have been pathologized by the medical profession, through classifications of mental illness in the Diagnostic and Statistical Manual of Mental Disorders (DSM) and International Classification of Disease (ICD). Although this is changing in the new ICD-11, tension remains between depathologization discourses and access to gender-affirming health care. Trans and gender diverse people experience significant health disparities and an increased burden of disease, specifically in the areas of mental health, Human Immunodeficiency Virus, violence and victimisation. Many of these health disparities originate from discrimination and systemic biases that decrease access to care, as well as from health professional ignorance. This paper will outline gaps in health science curricula that have been described in different contexts, and specific educational interventions that have attempted to improve awareness, knowledge and skills related to gender-affirming health care. The education of primary care providers is critical, as in much of the world, specialist services for gender-affirming health care are not widely available. The ethics of the gatekeeping model, where service providers decide who can access care, will be discussed and contrasted with the informed-consent model that upholds autonomy by empowering patients to make their own health care decisions. CONCLUSION: There is an ethical imperative for health professionals to reduce health care disparities of trans and gender diverse people and practice within the health care values of social justice and cultural humility. As health science educators, we have an ethical duty to include gender-affirming health in health science curricula in order to prevent harm to the trans and gender diverse patients that our students will provide care for in the future.

A multidimensional approach to inform family planning needs, preferences and behaviours amongst women in South Africa through body mapping.

BACKGROUND: In recent decades there have been great improvements in the reproductive health of women in low- and middle-income countries and increases in the use of modern contraceptive methods. Nonetheless, many women are not able to access information, contraceptive technologies and services that could facilitate preventing unintended pregnancies and planning the number and timing of desired pregnancies. In South Africa, the contraceptive prevalence rate is 64.6%. However, this relatively high contraceptive prevalence rate masks problems with quality contraceptive service delivery, equitable access, and women's ability to correctly and consistently, use contraceptive methods of their choice. This study set out to understand the specific family planning and contraceptive needs and behaviours of women of reproductive age in South Africa, through a lived experience, multisensory approach. METHODS: Participatory qualitative research methods were used including body mapping workshops amongst reproductive aged women recruited from urban and peri urban areas in the Western Cape South Africa. Data including body map images were analysed using a thematic analysis approach. RESULTS: Women had limited biomedical knowledge of the female reproductive anatomy, conception, fertility and how contraceptives worked, compounded by a lack of contraceptive counseling and support from health care providers. Women's preferences for different contraceptive methods were not based on a single, sensory or experiential factor. Rather, they were made up of a composite of sensory, physical, social and emotional experiences underscored by potential for threats to bodily harm. CONCLUSIONS: This study highlighted the need to address communication and knowledge gaps around the female reproductive anatomy, different contraceptive methods and how contraception works to prevent a pregnancy. Women, including younger women, identified sexual and reproductive health knowledge gaps themselves and identified these gaps as important factors that influenced uptake and effective contraceptive use. These knowledge gaps were overwhelmingly linked to poor or absent communication and counseling provided by health care providers. Body mapping techniques could be used in education and communication strategies around sexual and reproductive health programmes in diverse settings.

In silico design and optimization of selective membranolytic anticancer peptides.

Membranolytic anticancer peptides represent a potential strategy in the fight against cancer. However, our understanding of the underlying structure-activity relationships and the mechanisms driving their cell selectivity is still limited. We developed a computational approach as a step towards the rational design of potent and selective anticancer peptides. This machine learning model distinguishes between peptides with and without anticancer activity. This classifier was experimentally validated by synthesizing and testing a selection of 12 computationally generated peptides. In total, 83% of these predictions were correct. We then utilized an evolutionary molecular design algorithm to improve the peptide selectivity for cancer cells. This simulated molecular evolution process led to a five-fold selectivity increase with regard to human dermal microvascular endothelial cells and more than ten-fold improvement towards human erythrocytes. The results of the present study advocate for the applicability of machine learning models and evolutionary algorithms to design and optimize novel synthetic anticancer peptides with reduced hemolytic liability and increased cell-type selectivity.

Effects of 6 Weeks of Different High-Intensity Interval and Moderate Continuous Training on Aerobic and Anaerobic Performance.

Cavar, M, Marsic, T, Corluka, M, Culjak, Z, Cerkez Zovko, I, Müller, A, Tschakert, G, and Hofmann, P. Effects of 6 weeks of different high-intensity interval and moderate continuous training on aerobic and anaerobic performance. J Strength Cond Res 33(1): 44-56, 2019-To provide practical data, we compared the training effects of 3 different programs, using a shuttle run stimulus, on aerobic and anaerobic performance, measured using the 20-m maximal shuttle run (Beep) test and 300-yd shuttle run, respectively. Forty-five physically trained men, with a mean age of 21.1 ± 1.8 years, participated. The 6-week, 12-session training programs included 2 high-intensity interval training (HIIT) protocols, with either a short (SH) or long (LH) shuttle run interval, and a continuous shuttle run (CON), which was used as a control. The training intensity was based on the maximal shuttle run speed (MASS), measured on the Beep test, to elicit the relevant values of the time to exhaustion (TTE). Short (SH) training was performed at 115-120%(MASS), with a 10-second work to 10-second rest scheme, and the number of repetitions to be completed set to 70% of each participant's maximum (∼15 repetitions). LH training was performed at an intensity of 90-95%(MASS), with the duration set to 70%(TTE) (∼4 minutes). For both SH and LH, 3 sets were completed at each session, with a 2-3 minutes of rest between sets. CON training consisted of continuous shuttle running for 35 minutes at an intensity of 70%(MASS). Both SH and LH yielded a large training effect (p < 0.01), with SH preferentially improving anaerobic performance and LH preferentially improving aerobic performance. No effect of CON training was identified. Our findings indicate that these different training protocols cannot be used interchangeably and that the Beep test is useful in prescribing the intensity and duration of HIIT.

Missing knowledge of gendered power relations among non-governmental organisations doing right to health work: a case study from South Africa.

BACKGROUND: Despite 20 years of democracy, South Africa still suffers from profound health inequalities. Gender roles and norms are associated with individuals' vulnerability that lead to ill-health. For instance, gender inequality influences women's access to health care and women's agency to make health-related decisions. This paper explores gender-awareness and inclusivity in organisations that advocate for the right to health in South Africa, and analyses how this knowledge impacts their work? METHODS: In total, 10 in-depth interviews were conducted with members of The Learning Network for Health and Human Rights (LN), a network of universities and Civil Society Organisations (CSOs) which is explicitly committed to advancing the right to health, but not explicitly gendered in its orientation. RESULTS: The results show that there is a discrepancy in knowledge around gender and gendered power relations between LN members. This discrepancy in understanding gendered power relations suggests that gender is 'rendered invisible' within the LN, which impacts the way the LN advocates for the right to health. CONCLUSIONS: Even organizations that work on health rights of women might be unaware of the possibility of gender invisibility within their organisational structures.

Designing Anticancer Peptides by Constructive Machine Learning.

Constructive (generative) machine learning enables the automated generation of novel chemical structures without the need for explicit molecular design rules. This study presents the experimental application of such a deep machine learning model to design membranolytic anticancer peptides (ACPs) de novo. A recurrent neural network with long short-term memory cells was trained on α-helical cationic amphipathic peptide sequences and then fine-tuned with 26 known ACPs by transfer learning. This optimized model was used to generate unique and novel amino acid sequences. Twelve of the peptides were synthesized and tested for their activity on MCF7 human breast adenocarcinoma cells and selectivity against human erythrocytes. Ten of these peptides were active against cancer cells. Six of the active peptides killed MCF7 cancer cells without affecting human erythrocytes with at least threefold selectivity. These results advocate constructive machine learning for the automated design of peptides with desired biological activities.

The no-go zone: a qualitative study of access to sexual and reproductive health services for sexual and gender minority adolescents in Southern Africa.

BACKGROUND: Adolescents have significant sexual and reproductive health needs. However, complex legal frameworks, and social attitudes about adolescent sexuality, including the values of healthcare providers, govern adolescent access to sexual and reproductive health services. These laws and social attitudes are often antipathetic to sexual and gender minorities. Existing literature assumes that adolescents identify as heterosexual, and exclusively engage in (heteronormative) sexual activity with partners of the opposite sex/gender, so little is known about if and how the needs of sexual and gender minority adolescents are met. METHODS: In this article, we have analysed data from fifty in-depth qualitative interviews with representatives of organisations working with adolescents, sexual and gender minorities, and/or sexual and reproductive health and rights in Malawi, Mozambique, Namibia, Zambia and Zimbabwe. RESULTS: Sexual and gender minority adolescents in these countries experience double-marginalisation in pursuit of sexual and reproductive health services: as adolescents, they experience barriers to accessing LGBT organisations, who fear being painted as "homosexuality recruiters," whilst they are simultaneously excluded from heteronormative adolescent sexual and reproductive health services. Such barriers to services are equally attributable to the real and perceived criminalisation of consensual sexual behaviours between partners of the same sex/gender, regardless of their age. DISCUSSION/ CONCLUSION: The combination of laws which criminalise consensual same sex/gender activity and the social stigma towards sexual and gender minorities work to negate legal sexual and reproductive health services that may be provided. This is further compounded by age-related stigma regarding sexual activity amongst adolescents, effectively leaving sexual and gender minority adolescents without access to necessary information about their sexuality and sexual and reproductive health, and sexual and reproductive health services.

Recurrent Neural Network Model for Constructive Peptide Design.

We present a generative long short-term memory (LSTM) recurrent neural network (RNN) for combinatorial de novo peptide design. RNN models capture patterns in sequential data and generate new data instances from the learned context. Amino acid sequences represent a suitable input for these machine-learning models. Generative models trained on peptide sequences could therefore facilitate the design of bespoke peptide libraries. We trained RNNs with LSTM units on pattern recognition of helical antimicrobial peptides and used the resulting model for de novo sequence generation. Of these sequences, 82% were predicted to be active antimicrobial peptides compared to 65% of randomly sampled sequences with the same amino acid distribution as the training set. The generated sequences also lie closer to the training data than manually designed amphipathic helices. The results of this study showcase the ability of LSTM RNNs to construct new amino acid sequences within the applicability domain of the model and motivate their prospective application to peptide and protein design without the need for the exhaustive enumeration of sequence libraries.

Beyond 'invisibility': queer intelligibility and symbolic annihilation in healthcare.

Sexual minority health is increasingly receiving attention by health sciences education and healthcare, with the core argument being that health can be improved by challenging sexual minority invisibility. Invisibility as a concept, however, does not allow for a deeper theoretical engagement with the reasons and consequences of the lack of representation of queerness in healthcare. Drawing on empirical research with queer healthcare users in South Africa, I argue that 'invisibility' actually encompasses two distinct, though related, concepts: queer symbolic annihilation as the reason for the exclusion of queer identities in health professions education and, by consequence, in healthcare; and queer (un)intelligibility as the consequence of this systemic erasure. By simply attributing discriminatory healthcare experiences of queer people to 'invisibility' we are missing opportunities to address underlying issues of queer symbolic annihilation and unintelligibility.