Dimensionality reduction by UMAP reinforces sample heterogeneity analysis in bulk transcriptomic data.

Transcriptomic analysis plays a key role in biomedical research. Linear dimensionality reduction methods, especially principal-component analysis (PCA), are widely used in detecting sample-to-sample heterogeneity, while recently developed non-linear methods, such as t-distributed stochastic neighbor embedding (t-SNE) and uniform manifold approximation and projection (UMAP), can efficiently cluster heterogeneous samples in single-cell RNA sequencing analysis. Yet, the application of t-SNE and UMAP in bulk transcriptomic analysis and comparison with conventional methods have not been achieved. We compare four major dimensionality reduction methods (PCA, multidimensional scaling [MDS], t-SNE, and UMAP) in analyzing 71 large bulk transcriptomic datasets. UMAP is superior to PCA and MDS but shows some advantages over t-SNE in differentiating batch effects, identifying pre-defined biological groups, and revealing in-depth clusters in two-dimensional space. Importantly, UMAP generates sample clusters uncovering biological features and clinical meaning. We recommend deploying UMAP in visualizing and analyzing sizable bulk transcriptomic datasets to reinforce sample heterogeneity analysis.

EFL Students' Preferences for Written Corrective Feedback: Do Error Types, Language Proficiency, and Foreign Language Enjoyment Matter?

Using both quantitative and qualitative approaches, this study investigated the preference of learners of English as a foreign language (EFL) for four types of written corrective feedback (WCF), which are often discussed in the literature, on grammatical, lexical, orthographic, and pragmatic errors. In particular, it concerned whether such preference is influenced by two learner variables, namely, foreign language enjoyment (FLE) and proficiency level. The preference for selective vs. comprehensive WCF was also examined. The participants in the study were 117 University students in a Thai EFL context. Analysis of questionnaire data revealed a tendency for learners to prefer more explicit types of WCF (i.e., metalinguistic explanation and overt correction) for most error types, irrespective of their proficiency and FLE level. High proficiency level learners rated less explicit WCF types (i.e., underlining and error code) as useful to some degree, whereas their low proficiency level counterparts did not. Similar results were found for the two FLE groups. Besides, the FLE level seemed to play a role in perceiving the value of WCF in terms of scope. The results of follow-up interviews showed that the linguistic features of learners' first language, existing knowledge of the target language, affective feelings, and teacher's role were the main factors contributing to variation in learners' preferences. Possible pedagogical implications are discussed.