Searching journal content for articles similar to Li et al. 34 (9): 1421.

Displaying results 1-10 of 14
For checked items
  1. ...embeddings as input. Our gene470 embeddings are computed based on the pre-trained Enformer Avsec et al (2021). We471 tested other choices and they did not work as well as the setting based on Enformer.472 The loss is computed based on the measured genes between the raw input data and473 imputed outputs...
    OPEN ACCESS ARTICLEACCEPTED MANUSCRIPT
  2. ...and ground truth labels on a data set with n cells. NMI is defined as where I(U,G) represents the mutual information between U and G and is defined as and H(U) and H(G) are the entropies: AC is defined as the best matching between predicted and true clusters, which is given as where are the true labels...
  3. ...when data sets are analyzed in isolation. However, SpiceMix requires a common feature space, which limits its application to data sets with few overlapping genes. We address this limitation using LLOKI embeddings as an alternative input to SpiceMix, providing a unified feature space across data sets...
  4. ...transcriptomics (ST) data set, the spatial information L and the gene expression information E are integrated to produce a graph that contains both cell location and gene expression information. (B) A graph-embedding scheme converts a graph into a vector representation, as shown on the left, which is ready...
  5. ...representing any desired function. The model, now conditioned on the prompt tokens, generates a DNA sequence 1 nt at a time (Fig. 1D). In parallel, we train a supervised sequence-to-activity regression model on the same data set (Fig. 1E), and apply it to the generated sequences to select those that best match...
  6. ...the consequences of this design choice by comparing human WTA with a smaller probe pool targeting 1812 human genes with five probes per target. In matched 200 μm AOIs in formalin-fixed paraffin embedded (FFPE) tonsil tissue, counts from the single WTA probe were well correlated to the mean count of the five probes...
  7. ...that are homologous to Arabidopsis thaliana pericentromeres (peri- CEN s) III and V, and used FISH to demonstrate they have been maintained near centromere satellite arrays in each species. Sequence analysis revealed a set of highly conserved genes, yet we discovered substantial differences in intergenic length...
  8. ...Selenoprofiles to search for SelD/SPS genes in all available fully sequenced eukaryotic and prokaryotic s, 505 and 8263, respectively. We then utilized a combination of approaches to reconstruct their phylogenetic history. Methods and analyses are fully discussed in Supplemental Material S1–S6. SelD as a marker...
  9. ...cleavage. (A) Distribution of miRNA:mRNA alignment mismatch scores in Nematostella vectensis, Homo sapiens, Drosophila melanogaster, and Arabidopsis thaliana, shown as the difference in abundance between miRNAs and matching shuffledmiRNAs, counting the fraction of sequences in each set having a best match...
  10. ....1101/gr.279142.124 ↵Li L, Dannenfelser R, Cruz C, Yao V. 2024. A best-match approach for gene set analyses in embedding spaces. Genome Res (this issue) 34: 1421–1433. doi:10.1101/gr.279141.124 ↵Park C, Mani S, Beltran-Velez N, Maurer K, Huang T, Li S, Gohil S, Livak KJ, Knowles DA, Wu CJ, et al. 2024...
For checked items

Preprint Server