Average Linkage Clustering Research Articles

Hypergraph convolutional networks with multi-ordering relations for cross-document event coreference resolution

Recognizing the coreference relationship between different event mentions in the text (i.e., event coreference resolution) is an important task in natural language processing. It helps to understand the association between various events in the text, and plays an important role in information extraction, question answering systems, and reading comprehension. Existing research has made progress in improving the performance of event coreference resolution, but there are also some shortcomings. For example, most of the existing methods analyze the event data in the document in a serial processing mode, without considering the complex relationship between events, and it is difficult to mine the deep semantics of events. To solve these problems, this paper proposes a cross-document event co-reference resolution method (HGCN-ECR) based on hypergraph convolutional neural networks. Firstly, the BiLSTM-CRF model was used to label the semantic role of the events extracted from a number of documents. According to the labeling results, the trigger words and non-trigger words of the event were determined, and the multi-document event hypergraph was constructed around the event trigger words. Then hypergraph convolutional neural networks are used to learn higher-order semantic information in multi-document event hypergraphs, and multi-head attention mechanisms are introduced to understand the hidden features of different event relationship types by treating each event relationship as a set of separate attention mechanisms. Finally, the feed-forward neural network and the average link clustering method are used to calculate the coreference score of events and complete the coreference event clustering, and the cross-document event coreference resolution is realized. The experimental results show that the cross-document event co-reference resolution method is superior to the baseline model.

AVERAGE LINKAGE CLUSTERING METHOD AND MOLECULAR DOCKING STUDY ON DATE PALM (PHOENIX DACTYLIFERA L.) AS POTENTIAL ANTI-ANEMIA AGENT

Anemia, characterized by blood hemoglobin (Hb) levels below the World Health Organization's (WHO) normal limit, remains a significant health concern. Date fruit (Phoenix dactylifera L.) stands out as an herbal plant boasting the highest iron content at 13.7 mg, suggesting its potential as an anti-anemia agent. This study aimed to explore the anti-anemia potential of active compounds in date fruit using average linkage clustering and validated using molecular docking. Compounds from dates were gathered via GC-MS analysis and online databases, totaling 145 compounds—50 from GC-MS and 95 from Knapsack and Dr. Duke databases. Additionally, 5 lead compounds served as positive controls for comparison. SwissADME online servers assessed the compounds' properties, serving as materials for the clustering method. The average linkage clustering method was employed, yielding an excellent dendrogram with a cophenetic correlation of 0.711. Notably, a total of 17 date fruit compounds are in the same cluster as the lead compounds. Molecular docking revealed 4 date palm fruit-derived compounds as potential PHD enzyme inhibitors, promising for anemia treatment. In conclusion, the average linkage clustering method and validation using molecular docking approaches highlight date fruit's potential as an alternative anemia treatment, showcasing the significance of interdisciplinary methodologies in drug discovery.

Recovering nested structures in networks: an evaluation of hierarchical clustering techniques

Abstract In this article, we present various algorithms to partition the nodes of a network into groups that show the property of nestedness. Since perfect nestedness is a rare phenomenon, we consider the task from a data mining perspective, and we search for groups having high-level of nestedness. We utilize both agglomerative and divisive hierarchical clustering procedures and compare them on several benchmark and real-life networks. Furthermore, we propose different metrics derived from the results of our algorithms. We show that average-linkage and complete-linkage clustering can recover the largest fully nested clusters, and that the cluster size-weighted mean nestedness was a more stable metric for measuring clustering performance. Our proposed algorithms allow us to create multiple resolution views of nestedness-based clustering of networks, extending the field of graph-based data mining.

Volatile and non-volatile metabolite profiling of under-explored crop Piper colubrinum link. With chromatography mass spectrometry approach and its biochemical diversity from medicinally valued Piper species

Piper colubrinum Link. is an underexplored crop regarding its metabolites and therapeutic attributes. Current study aimed to identify the possible volatile and non-volatile metabolites of P. colubrinum fruit and studied its metabolite diversity with medicinally valued Piper species viz. P. nigrum L., P. longum L. and P. chaba Hunter. The volatile constituents of P. colubrinum essential oil by GC–MS revealed the presence of sesquiterpenes as the major contribution. The sesquiterpenes α-muurolol (12.5 %) and β-caryophyllene (11.3 %) were the predominant volatile components. Few aliphatic compounds like n-heptadecane and trace amounts of monoterpenes (α- and β-pinene and α-terpineol) were also identified from this crop. The fatty acid profiling by GC–MS revealed mainly oleic acid (41.3 %) followed by palmitic and linoleic acids. HPLC analysis demonstrated that the major pungent alkaloid piperine was found to be trace (0.04 %) in P. colubrinum. The LC-QTOF-MS/MS profiling of the chloroform extract of the P. colubrinum revealed the presence of non-volatile constituents including phenolic and alkaloid compounds. Ferulic acid, rosmarinic acid, salicylic acid, kaempferol-5-glucoside, 5-methoxysalicylic acid, apigenin-7-galactoside, kaempferide-3-glucoside, luteolin, kaempferol, apigenin and scutellarein-4′-methyl ether were the phenolic compounds whereas piperlonguminine was the alkaloid compound identified. Finally, the biochemical parameters of this crop were compared with that of P. nigrum, P. longum and P. chaba and average linkage cluster dendrogram revealed that P. colubrinum was biochemically distinct from other three Piper species.

Clustering analysis for the evolutionary relationships of SARS-CoV-2 strains.

To explore the differences and relationships between the available SARS-CoV-2 strains and predict the potential evolutionary direction of these strains, we employ the hierarchical clustering analysis to investigate the evolutionary relationships between the SARS-CoV-2 strains utilizing the genomic sequences collected in China till January 7, 2023. We encode the sequences of the existing SARS-CoV-2 strains into numerical data through k-mer algorithm, then propose four methods to select the representative sample from each type of strains to comprise the dataset for clustering analysis. Three hierarchical clustering algorithms named Ward-Euclidean, Ward-Jaccard, and Average-Euclidean are introduced through combing the Euclidean and Jaccard distance with the Ward and Average linkage clustering algorithms embedded in the OriginPro software. Experimental results reveal that BF.28, BE.1.1.1, BA.5.3, and BA.5.6.4 strains exhibit distinct characteristics which are not observed in other types of SARS-CoV-2 strains, suggesting their being the majority potential sources which the future SARS-CoV-2 strains' evolution from. Moreover, BA.2.75, CH.1.1, BA.2, BA.5.1.3, BF.7, and B.1.1.214 strains demonstrate enhanced abilities in terms of immune evasion, transmissibility, and pathogenicity. Hence, closely monitoring the evolutionary trends of these strains is crucial to mitigate their impact on public health and society as far as possible.

Morphological Diversity and Genetic Parameter of Okra (Abelmoschus esculentus L. Moench) Genotypes.

<b>Background and Objective:</b> Considering that the potential for okra as an anti-diabetic is very high, while okra productivity in Indonesia is still low, a plant breeding program through variety development is needed. One of the initial activities that needs to be carried out is the characterization of various genotypes, both quantitative and qualitative characters. This research aimed to obtain information on the diversity of morpho-agronomic characters in okra genotypes. <b>Materials and Methods:</b> The experiment was conducted as a randomized block design, one factor is genotype with three replications. The materials used in this research were 20 okra genotypes. The experimental units used in this research were 60 units. Each experimental unit consists of 10 sample plants. Analysis of quantitative character variations used PKBT-STAT 3.1. Cluster analysis was carried out with PBSTAT-CL 2.1.2 with the Gower dissimilarity and average linkage clustering methods. Furthermore, analysis was carried out using SAS OnDemand for Academics to see the distinguishing characteristics between clusters. <b>Results:</b> There were differences in okra genotypes based on qualitative and quantitative characteristics. The most diverse quantitative character is the yield component, which is the fruit character. Variance in genetic and heritability showed broad and high criteria, respectively. Based on cluster analysis results, okra genotypes were grouped into 3 clusters with a cophenetic distance value of 0.40. Cluster 1 consists of 9 genotypes. Cluster 2 consists of 10 genotypes. Cluster 3 consists of 1 genotype the Red Hill Country genotype. The grouping in cluster analysis was carried out based on leaf width, number of fruits, fruit weight, fruit diameter and carpel thickness character. <b>Conclusion:</b> This diversity of okra germplasm can facilitate plant breeding activities in the future by selecting genotypes to serve as parents according to the objectives carried out.

ANALYSIS OF TRADITIONAL CHINESE MEDICINE PRESCRIPTIONS FOR CEREBRAL STROKE-RELATED DISEASES

Cerebral stroke, a type of cerebrovascular disease, has become the second leading cause of death globally. It is closely related to many diseases, including hypertension, diabetes, senile dementia, and so forth. As traditional Chinese medicine formulas are increasingly used to treat stroke and its associated diseases, people have begun to use machine learning methods to analyze Chinese medicine prescriptions and summarize their laws. In this study, we collected the data from classic Chinese formulations. Using the Jaccard similarity coefficient method, we calculated the similarity between different prescriptions. We then employed average linkage clustering, categorizing medicine prescriptions for chronic diseases such as diabetes, hypertension, and coronary heart disease into 12 groups. Some of these included “Heart Failure and Warm Kidney Soup”, “Sinus Chamber Junction Syndrome, Quadruple One Depression, Yi Qi, Activating Blood and Feeding Heart Soup”, “Nerves One, Depression One, Tian Ma Hook and Vine Drink”, and “Bawei Antihypertensive Decoction, Anemia Decoction, Hypotension Decoction, Myocardial Live Drink”. We observed that similar prescriptions had more meaningful mutual references. Subsequently, a correlation algorithm was used to analyze the “indications” and “prescription composition”, revealing 11 effective correlation rules. Among these, palpitations were strongly correlated with Astragalus membranaceus, Angelica sinensis, and cassia twig; weakness with Salvia miltiorrhiza, A. membranaceus, and A. sinensis; headaches with Ligusticum wallichii; and vertigo with A. membranaceus. These findings provided a theoretical reference for using traditional Chinese medicine in treating cerebral stroke and associated illnesses.

Multivariate Analysis for Post Harvest Quality and Yield Attributes in Tomato (Solanum lycopersicum L.)

Aim: To study the multivariate analysis, genetic parameters and correlation for post harvest quality and yield traits in tomato.
 Study Design: The variability in the twenty genotypes of tomato (Solanum lycopersicum L.) was evaluated for 18 yield attributes and post harvest quality traits using randomized block design and analyzed with multivariate methods.
 Place and Duration of Study: Twenty genotypes of tomato augmented from Indian Institute of Vegetable Research, Varanasi were sown during rainy season at Horticulture Research Farm of Banaras Hindu University.
 Methodology: The unweighted pair group method of the average linkage (UPGMA) cluster analysis and principal component analysis (PCA) were used to analyze the data. Canonical discriminant analysis showed the contribution of each trait to the classification of the tomato accessions into different cluster groups.
 Results: The first seven principal components (PC) explained 87.83% of total variation and has eigen values >1. The traits that mainly contributed for this variation in PC1 and PC2 are fruit yield, plant height, number of flower clusters per plant, number of fruits per plant, days to first fruit set, number of fruits per cluster. High (>20%) genotypic coefficient of variation (GCV) and phenotypic coefficient of variation (PCV) was observed for number of fruits/cluster, fruit shape index, number of fruits/plant, fruit yield and shelf life. The correlation coefficients of number of primary branches/plant, days to first fruit set, number of flower clusters/plant, number of fruits/cluster, fruit width (cm), number of fruits/plant, fruit weight (g) and lycopene (mg/100 g) were positively and significantly correlated to fruit yield/plant.
 Conclusion: Number of flower clusters/plant, fruit weight and number of fruits per plant had contributed for maximum variation. These traits also had high heritability, high genetic gain and significant correlation with fruit yield.

Selection of an adequate functional diversity index for stream assessment based on biological traits of macroinvertebrates

Functional diversity (FD) is useful for the evaluation of freshwater ecosystems. The FD of macroinvertebrate families for river water quality (WQ) assessment in the Paute River Basin (PRB), Ecuador, was investigated. Macroinvertebrate samples and data about 26 physical, chemical, microbiological and hydro-geomorphological variables were available. Literature-based biological traits were allocated as scores to the macroinvertebrates data through fuzzy coding. The Generalised Additive Mixed Model (GAMM) was used to assess the performance of six FD indices using the referred 26 WQ descriptive variables. The best performing GAMM led to selecting the index based on functional dendrograms including the species community pool (wFDc) as the most suitable to characterise FD in the PRB. The sub-basins of the PRB were grouped in three classes applying Average Linkage Clustering (ALC) and using wFDc. The Random Forest (RF) algorithm was used with a global efficiency of 89% to assess the ALC clusters consistency and pre-identify the significant WQ descriptive variables, explaining most of the FD variability. The Kruskal-Wallis test was then applied to refine the outcomes of the previous analysis. Twelve WQ descriptive variables were finally identified as the best discriminant predictors for FD, including the riparian vegetation, electric conductivity, dissolved oxygen, total hardness, faecal coliforms and pH. It is believed that the implemented approach successfully assessed the stream WQ status of the PRB upon selecting a suitable macroinvertebrate FD index; as such, it could be applied to other tropical basins for WQ assessment.

Combining performance and estimated genetic diversity among soybean parents and F1 populations

Studies about the combining ability and genetic diversity can provide information on the genetic resources contained in the working collection of germplasm. The objective of this study was to estimate the combining ability of ten soybean parents and their hybrids in the F1 generation using diallel analysis and genetic diversity analysis to evaluate the dissimilarity between parents and crosses. The experiment was carried out in a greenhouse, in a completely randomized design with five replications, during the 2019/2020 crop season, in the municipality of Rio Verde, State of Goiás, Brazil. Agronomic and physiological variables were evaluated. Data were subjected to analysis of variance. Diallel analysis was performed using the Griffing method 2. Analysis of genetic diversity was performed with the Mahalanobis generalized distance and the hierarchical average linkage clustering. Additive gene effects were predominant in the control of all variables analyzed in the F1 generation. The parents SYN 1163 RR and CG 7370 RR showed good complementarity for the variable grain yield. The thirty-five genotypes were separated into six groups, and the variable weight of one hundred seeds had the greatest contribution to genetic diversity.

3PNMF-MKL: A non-negative matrix factorization-based multiple kernel learning method for multi-modal data integration and its application to gene signature detection.

In this current era, biomedical big data handling is a challenging task. Interestingly, the integration of multi-modal data, followed by significant feature mining (gene signature detection), becomes a daunting task. Remembering this, here, we proposed a novel framework, namely, three-factor penalized, non-negative matrix factorization-based multiple kernel learning with soft margin hinge loss (3PNMF-MKL) for multi-modal data integration, followed by gene signature detection. In brief, limma, employing the empirical Bayes statistics, was initially applied to each individual molecular profile, and the statistically significant features were extracted, which was followed by the three-factor penalized non-negative matrix factorization method used for data/matrix fusion using the reduced feature sets. Multiple kernel learning models with soft margin hinge loss had been deployed to estimate average accuracy scores and the area under the curve (AUC). Gene modules had been identified by the consecutive analysis of average linkage clustering and dynamic tree cut. The best module containing the highest correlation was considered the potential gene signature. We utilized an acute myeloid leukemia cancer dataset from The Cancer Genome Atlas (TCGA) repository containing five molecular profiles. Our algorithm generated a 50-gene signature that achieved a high classification AUC score (viz., 0.827). We explored the functions of signature genes using pathway and Gene Ontology (GO) databases. Our method outperformed the state-of-the-art methods in terms of computing AUC. Furthermore, we included some comparative studies with other related methods to enhance the acceptability of our method. Finally, it can be notified that our algorithm can be applied to any multi-modal dataset for data integration, followed by gene module discovery.

Developing the New Thermal Climate Zones of China for Building Energy Efficiency Using the Cluster Approach

Climate is a key element in building design. The determination of adequate climate zoning is crucial for achieving building energy efficiency and reduced carbon emissions. In this study, a cluster analysis method was applied to develop new thermal climate zones for building envelope thermal design in China. In total, three different cluster analysis methods, including k-means, average-linkage, and Ward’s clustering, were considered. The analysis indicated that the average-linkage clustering was more appropriate for this study, and the elbow method could not accurately determine the best number of categories of average-linkage clustering. Further analysis showed that the unsupervised cluster processes may generate an unavoidable redundancy category and, to obtain precise results, supervision may be necessary in some contexts. Finally, China was classified into 10 climate zones. The North China plain and Qinghai–Tibet plateau are classified into two independent climate zones, and Turpan and Bayanbulak were classified into two newly defined climate zones different from their surrounding area in the new definition. Quantitative analysis indicated that the new zones were more precise when compared to the current ones, which can provide more precise climate information and contribute to formulating more precise standards and policies related to the thermal design of building envelopes in the future.

Intraspecific Diversity in Aquatic Ecosystems: Comparison between Spirodela polyrhiza and Lemna minor in Natural Populations of Duckweed.

Samples of two duckweed species, Spirodela polyrhiza and Lemna minor, were collected around small ponds and investigated concerning the question of whether natural populations of duckweeds constitute a single clone, or whether clonal diversity exists. Amplified fragment length polymorphism was used as a molecular method to distinguish clones of the same species. Possible intraspecific diversity was evaluated by average-linkage clustering. The main criterion to distinguish one clone from another was the 95% significance level of the Jaccard dissimilarity index for replicated samples. Within natural populations of L. minor, significant intraspecific genetic differences were detected. In each of the three small ponds harbouring populations of L. minor, based on twelve samples, between four and nine distinct clones were detected. Natural populations of L. minor consist of a mixture of several clones representing intraspecific biodiversity in an aquatic ecosystem. Moreover, identical distinct clones were discovered in more than one pond, located at a distance of 1 km and 2.4 km from each other. Evidently, fronds of L. minor were transported between these different ponds. The genetic differences for S. polyrhiza, however, were below the error-threshold of the method within a pond to detect distinct clones, but were pronounced between samples of two different ponds.

Decision-making in formation of mean-VaR optimal portfolio by selecting stocks using K-means and average linkage clustering

Stock is one of the investment assets that has its charm for investors. It is very liquid and has a high rate of return, but it has a high risk. The strategy commonly used to minimize investment risk is to diversify through portfolio formation. A good allocation of funds must be determined in forming an optimal portfolio. In addition, the method of stock selection needs to be considered so the stocks are well diversified and the portfolio developed has good performance. This study aims to compare stock selection between K-Means and Average Linkage clustering approaches in forming an investment portfolio. Clustering analysis is used to group IDX80 stocks based on their attributes. In forming a portfolio with the Mean-VaR model, the stock selection decision criteria used are by selecting stocks with the highest positive returns from each cluster. As a result, the two clustering techniques show the superiority of the Silhouette score for a certain number of clusters, but there are still more advantages in Average Linkage. The portfolio approached by Average Linkage resulted in a better performance than the portfolio approached by K-Means. Therefore, Average Linkage clustering can be used as a better recommendation in decision-making to select stocks so as to produce optimal portfolio performance.

Diversity of milk traits in 25 goat breeds

Goats are livestock that is mostly raised by small farmers in Indonesia because they are easier to raise. Apart from having the potential to become meat, several breeds of which are kept as milk-purpose. Milk traits of each breed differ from one another. Therefore, this study tried to observe genetic differences of 25 goat breeds with statistical approach. Information about milk traits from 25 goat breeds is obtained from published journal. Multidimensional preference analysis and average linkage cluster analysis were performed using SAS 9.4 to determine the differences in goat breeds from three traits, namely milk yield, fat content, and protein content. Multidimensional preference analysis was able to see the advantages of breeds from the three observed traits. Goat breeds with superior milk yields are Saanen, Camosciata delle Alpi, and Charmoisée. Sarda Primitiva, Sarda, Etawah Grade have high fat content in milk. In addition, Arsi-Bale and Somali have high protein content. Average linkage cluster analysis is able to observe the genetic relationship of goat breeds based on three traits. According to average linkage cluster analysis, we found four clusters for goat breeds in this study. With existing statistical approaches, we can evaluate genetic diversity in milk traits.

Multivariate Analysis of Butterfly Pea (Clitoria ternatea L.) Genotypes With Potentially Healthy Nutraceuticals and Uses

Butterfly pea (Clitoria ternatea L.) is a legume used as tea, forage, ornamental, salad, and medicinal plant. The flowers range from white to dark purple with little known about the variation for seed and flower color in the United States Department of Agriculture, Agricultural Research Service, Plant Genetic Resources Conservation Unit germplasm collection. Therefore, 26 butterfly pea accessions were analyzed using a principal component analysis (PCA) and average linkage cluster analysis (ALCA). These butterfly pea genotypes ranged from 56% to 99% for viabilities, 2.57 to 5.88 g for 100 seed weight, 34.07 to 226.26 g for total seed weight, and 1,326 to 3,874 for total seed numbers. PCA accounted for 40%, 57%, 70%, 79%, and 86% of the variation using principal components (PCs) 1 through 5, respectively. PC1 was most correlated with 100 and total seed weight, while PC2 correlated with blue, white, and purple flowers. PC3 correlated mostly with germination, purple flowers, and total seed weight. PCs 4 and 5 primarily correlated with blue and purple flowers, respectively. Several significant correlations were also observed. ALCA grouped the 26 butterfly pea genotypes into four distinct seed number–producing clusters. Clusters 1 to 4 represent the lowest to highest seed numbers produced by the butterfly pea genotypes. Several potential health benefits from butterfly pea flowers, leaves, seeds, and roots for human use were identified from the literature.

Comparison of Microsatellite DNA Fingerprints of Original Seed Files and Conserved Rice Germplasm Collections

The understanding of the genetic identity and composition of germplasm collections conserved in a genebank is vital for the proper management of germplasm resources and utilization in various plant breeding programs. A study was conducted at the Genebank of the Philippine Rice Research Institute (PhilRice) to examine the genetic similarity of the original seed files (SF) and regenerated seeds conserved as an active collection (AC) using 18 microsatellite DNA markers and to explore their underlying intra-varietal heterogeneity. Here, 74 rice germplasm collections regenerated from 1992–2012 were randomly selected for the comparative DNA fingerprinting. The genetic similarity between the original SF and regenerated AC was examined through the Dice coefficient (DC) analysis, while the extent of intra-varietal genetic heterogeneity was determined via average-linkage cluster analysis based on binary genetic distance. Of the 74 collections studied, 68 showed paired clustering of SF and AC, indicating high genetic similarity. The DCs of these collections ranged from 0.79–1. On the other hand, SF and AC comparison of six germplasm collections yielded very low DCs ranging from 0.62–0.77, with their corresponding SF and AC pair dissociated in the dendrogram owing to their high intra-varietal genetic heterogeneity. The intra-varietal polymorphic alleles between SF and AC were observed in all markers analyzed – especially in RM562, RM547, and RM154 microsatellite markers –suggesting their potential use in assessing the levels of intra-varietal genetic heterogeneity in rice cultivars. This pilot study provides baseline information that may be used for potential integration of intra-varietal similarity or variability tests in the Genebank which are important aspects for effective germplasm conservation, management, and utilization. Also discussed in this paper are other insights regarding the intra-varietal genetic heterogeneity of the selected rice accessions.

A new validity clustering index-based on finding new centroid positions using the mean of clustered data to determine the optimum number of clusters

Clustering, an unsupervised pattern classification method, plays an important role in identifying input dataset structures. It partitions input datasets into clusters or groups where either the optimum number of clusters is known in prior or automatically determined. In the case of automatic clustering, the performance is evaluated using a cluster validity index (CVI), which determines the optimum number of clusters in the data. From previous works, the improper cluster centroids positioning produced by clustering algorithms could reduce the performance of the validation process and performance produced by the previous state-of-the-art CVIs. In addition, those previous CVIs can only work properly with certain clustering algorithms and simple datasets structures, which their performances will reduce if they are applied to other clustering algorithms as well as more complex datasets. This study proposes an efficient CVI, namely, the validity clustering index based on finding the mean of clustered data (VCIM). The proposed approach combines the properties of the score function index and the mean to determine new cluster centroid positions. The performance of the VCIM index is compared with well-known CVIs on both artificial and real-life datasets. The obtained results on artificial datasets show that the proposed VCIM index outperforms the other CVIs in determining the true number of clusters for the five conventional clustering algorithms, namely, K-means, Fuzzy C-mean, agglomerative hierarchical average linkage clustering, variance-based differential evolution, and density peaks clustering and Particle swarm optimization (PDPC) algorithms. For the 14 real-word datasets, the proposed VCIM index correctly determined the optimum number of clusters for 11 out of 14 for the K-means clustering algorithm, 9 out of 14 for both Fuzzy clustering and agglomerative hierarchical average linkage clustering algorithms, 12 out of 14 for the variance-based differential evolution algorithm and 11 out of 14 datasets for PDPC. The obtained results using the proposed VCIM show its significance when combined with clustering algorithms and nominate its potential in various clustering applications.

P–769 FTIR spectroscopy analysis reveals differences between human embryo culture media composition by type of formulation and by manufacturer

Abstract Study question Can we detect the variation in different commercial human embryo culture media composition by Fourier-transform infrared (FTIR) spectroscopic analysis? Summary answer The spectra reveals distint features that allow distinguishing between continuous and sequential media, as well as between manufacturers of the same media type. What is known already: Manufacturers do not fully disclose commercially available culture media formulations. For this reason, it is important to gain insight into de differences between the available formulations, in order to understand how they might be linked with efficacy and safety ART parameters. Fourier Transform Infra-Red (FTIR) can be used for this purpose. Study design, size, duration Culture media samples (1 mL) were collected from local IVF laboratories, and stored frozen at –20 °C. Five repeats of 25 µL aliquots from the samples were analysed using FTIR spectroscopy to acquire the whole molecular fingerprint of each culture media. Participants/materials, setting, methods Three continuous (SAGE 1-STEP, Origio, G-TL, Vitrolife, and GERI, Genea) and four sequential (G1 PLUS, Vitrolife; Sequential Cleav with phenolred, Origio; G2 PLUS, Vitrolife and Sequential Blast with phenolred, Origio) media were analysed. Different pre-processing methods (atmospheric and baseline correction, normalizations, and derivatives) were carried out to minimize physical artefacts while highlighting chemical features. To compare the spectra of different media, multivariate analysis, as principal component analysis (PCA) and hierarchical cluster analysis (HCA) were employed. Main results and the role of chance The whole molecular fingerprint of all media analysed showed a similar pattern, revealing that, overall, the composition is very similar. However, PCA and HCA analysis revealed that significant differences exist, both between media type (continuous vs. sequential), and between different manufacturers within the same media type. Average linkage clustering using Spearman’s rank correlation confirms the similarities between the continuous and the sequential formulations. An analysis focusing the fingerprint region of the spectra (900 – 1700 cm^–1), also revealed variability between manufacturer, between media type (continuous vs. sequential) and of stage-specific media (cleavage vs. blastocyst). For instance, GERI media visually appeared to have distinct peaks compared to all other media, which was confirmed later through multivariate statistical analysis. Limitations, reasons for caution FTIR spectroscopy does not allow for a direct identification of the analytes present in the culture media, we can only infer the functional groups, but that are common on diverse biomolecules. Wider implications of the findings: FTIR analysis reveals differences between different media, such as cleavage and blastocyst specific, sequential and continuous, or manufacturer’s formulations. The increased resolution of the FTIR profile proves to be a powerful tool for analysing human embryo media, and could be used to establish correlations with media clinical performance and safety. Trial registration number Not applicable

Medication burden and clustering in people living with HIV undergoing therapeutic drug monitoring.

People living with HIV (PLWH) have a high burden of comorbidities and concomitant medication use. The aim of this study was to analyse the prevalence, predictors and patterns of polypharmacy (PP) in a large therapeutic drug monitoring (TDM) registry. We searched our TDM registry and categorized co-medications into 26 drug classes. We included patients with at least one medication recorded: PP and severe polypharmacy (sPP) were defined as the concomitant use of ≥5 or ≥10 nonantiretroviral/nonantitubercular drugs. Multivariable binary logistic analysis were conducted for identifying PP/sPP predictors. A hierarchical average-linkage cluster analysis was performed among drug classes. We included 2432 participants (1158 PLWH) aged 49.6 years (± 14.4) in the 2016-2020 period. A higher number of concomitant medications (4 vs 3.1, P < .001) and a higher prevalence of PP (26.1% vs 21.8%, P = .015) were recorded in controls. At multivariable binary logistic analysis older age, female gender and HIV-positive serostatus (P = .015) were independent predictors of PP; older age and year of inclusion were independent predictors of sPP. Cluster analysis showed that patients receiving oral drug for type 2 diabetes have a high probability of receiving several other drugs; a cluster of co-medications was observed with opioids, diuretics and central nervous system-affecting drugs. We observed a moderately high prevalence of polypharmacy in middle-aged PLWH: advanced age and female gender were associated with the greatest prevalence. The observation of co-medication clusters suggests groups of comorbidities but also identifies groups of patients at risk of similar drug-to-drug interactions.