Machine Learning Strategies for Improved Phenotype Prediction in Underrepresented Populations.

Precision medicine models often perform better for populations of European ancestry due to the over-representation of this group in the genomic datasets and large-scale biobanks from which the models are constructed. As a result, prediction models may misrepresent or provide less accurate treatment recommendations for underrepresented populations, contributing to health disparities. This study introduces an adaptable machine learning toolkit that integrates multiple existing methodologies and novel techniques to enhance the prediction accuracy for underrepresented populations in genomic datasets. By leveraging machine learning techniques, including gradient boosting and automated methods, coupled with novel population-conditional re-sampling techniques, our method significantly improves the phenotypic prediction from single nucleotide polymorphism (SNP) data for diverse populations. We evaluate our approach using the UK Biobank, which is composed primarily of British individuals with European ancestry, and a minority representation of groups with Asian and African ancestry. Performance metrics demonstrate substantial improvements in phenotype prediction for underrepresented groups, achieving prediction accuracy comparable to that of the majority group. This approach represents a significant step towards improving prediction accuracy amidst current dataset diversity challenges. By integrating a tailored pipeline, our approach fosters more equitable validity and utility of statistical genetics methods, paving the way for more inclusive models and outcomes.

Machine Learning Strategies for Improved Phenotype Prediction in Underrepresented Populations.

Precision medicine models often perform better for populations of European ancestry due to the over-representation of this group in the genomic datasets and large-scale biobanks from which the models are constructed. As a result, prediction models may misrepresent or provide less accurate treatment recommendations for underrepresented populations, contributing to health disparities. This study introduces an adaptable machine learning toolkit that integrates multiple existing methodologies and novel techniques to enhance the prediction accuracy for underrepresented populations in genomic datasets. By leveraging machine learning techniques, including gradient boosting and automated methods, coupled with novel population-conditional re-sampling techniques, our method significantly improves the phenotypic prediction from single nucleotide polymorphism (SNP) data for diverse populations. We evaluate our approach using the UK Biobank, which is composed primarily of British individuals with European ancestry, and a minority representation of groups with Asian and African ancestry. Performance metrics demonstrate substantial improvements in phenotype prediction for underrepresented groups, achieving prediction accuracy comparable to that of the majority group. This approach represents a significant step towards improving prediction accuracy amidst current dataset diversity challenges. By integrating a tailored pipeline, our approach fosters more equitable validity and utility of statistical genetics methods, paving the way for more inclusive models and outcomes.

Image recovery from unknown network mechanisms for DNA sequencing-based microscopy.

Imaging-by-sequencing methods are an emerging alternative to conventional optical micro- or nanoscale imaging. In these methods, molecular networks form through proximity-dependent association between DNA molecules carrying random sequence identifiers. DNA strands record pairwise associations such that network structure may be recovered by sequencing which, in turn, reveals the underlying spatial relationships between molecules comprising the network. Determining the computational reconstruction strategy that makes the best use of the information (in terms of spatial localization accuracy, robustness to noise, and scalability) in these networks is an open problem. We present a graph-based technique for reconstructing a diversity of molecular network classes in 2 and 3 dimensions without prior knowledge of their fundamental generation mechanisms. The model achieves robustness by obtaining an unsupervised sampling of local and global network structure using random walks, making use of minimal prior assumptions. Images are recovered from networks in two stages of dimensionality reduction first with a structural discovery step followed by a manifold learning step. By breaking the process into stages, computational complexity could be reduced leading to fast and accurate performance. Our method represents a means by which diverse molecular network generation scenarios can be unified with a common reconstruction framework.

Enzyme replacement therapy for the treatment of Hunter disease: A systematic review with narrative synthesis and meta-analysis.

BACKGROUND: In the last 10 years enzyme replacement therapy (ERT) has become an alternative for the treatment of patients with Hunter disease (HD). Nevertheless, the information regarding efficacy and safety is scarce and mainly based on the pivotal trials. This scarcity is especially evident for adults and severe forms of HD. METHODS: A systematic review of publications in the electronic databases PUBMED, EMBASE and Cochrane Central was undertaken. Clinical trials and observational studies were included. The data about efficacy and security were retrieved and analysed with Review Manager version 5.3. RESULTS: 677 records were found, 559 remaining after the removal of duplicates. By title and abstract review, 427 were excluded. Full reading of the rest was made (122 publications) and 42 were finally included. It was not possible to perform meta-analysis of all the endpoints due to high heterogeneity in the reporting and measuring of variables in each publication. Eight clinical trials were included, 6 with high risk of bias. The quality of the other studies was low in 12%, average in 68% and good in 21%. Main findings were: a reduction in the elimination of glycosaminoglycans (GAG) in urine in all the studies (26/26), decrease in liver and spleen size (18/18), increase of 52.59 m (95% CI, 36, 42-68.76, p < .001) in the 6-min walk test (TM6M), increase in forced vital capacity (FVC) of 9.59% (95% CI 4.77-14.51, p < .001), reduction of the left ventricular mass index of 3.57% (95% CI 1.2-5.93) and reduction in mortality (OR) of 0.44 (0.27-0.71). DISCUSSION: The data suggests a clear and consistent effect of ERT in HD reducing the accumulation of GAGs in the body, demonstrated by the reduction of its urinary excretion, as well as by the reduction of its deposits (spleen, liver and heart). Likewise, there is an improvement in physical and respiratory function. In addition, a reduction in mortality has been observed. Lack of studies, small size of the samples, and methodological deficiencies are the main limitations to establish definite conclusions. CONCLUSIONS: The data suggests that ERT is effective and safe in the treatment of HD. There is a need to evaluate patient-centred outcomes and the impact on quality of life.

Pseudohemorragia subaracnoidea. Presentación de un caso / Pseudo-subarachnoid Hemorrhage. A Case Report

La pseudohemorragia subaracnoidea es un fenómeno infrecuente que se caracteriza por hallazgos sugestivos de hemorragia subaracnoidea en la tomografía computarizada simple de cráneo, sin evidencia de la misma en estudios adicionales. Se ha asociado a múltiples causas, de las cuales la principal es la encefalopatía hipóxico-isquémica posparo cardiaco y reanimación cardiopulmonar. El contexto clínico y los niveles de atenuación medidos en Unidades Hounsfield (UH) se deben tener en cuenta al hacer el diagnóstico diferencial entre ambas entidades. Se presenta el caso de una paciente con pseudohemorragia subaracnoidea de etiología multifactorial.

Pseudo-subarachnoid hemorrhage (PSAH) is an infrequent entity characterized by findings in non- contrast head computed tomography that mimic subarachnoid hemorrhage, but without evidence of blood products in further studies. It has been associated with multiple etiologies, with hypoxic ischemic encephalopathy following cardiac arrest and cardiopulmonary resuscitation as the leading cause in literature. Clinical context and attenuation levels measured in Hounsfield Units should be taken into consideration when establishing the differential diagnosis between these entities. The case of a patient with PSAH of multifactorial etiology is presented.