bsc 1  (ATCC)

Journal: bioRxiv

Article Title: A genome language model for mapping DNA replication origins

doi: 10.64898/2026.01.29.702604

Figure Lengend Snippet: a) Heatmaps of Spearman correlation across combinations of probability- and efficiency-smoothing windows in H1, illustrating genome-wide sensitivity to resolution matching. b) Dependence of the Spearman correlation on smoothing-window size for multiple species/cell types (H1, mESC, and sheep fibroblast), plotted as mean ± standard deviation across autosomes, highlighting cross-species consistency.

Article Snippet: Chicken embryonic fibroblast cells (UMNSAH/DF-1) were purchased from ATCC (ATCC-CRL-3586) and were maintained according to the supplier’s instructions in DMEM (ATCC 30-2002) supplemented with 10% fetal bovine serum (Gibco, 10270-106).

Techniques: Genome Wide, Standard Deviation

Journal: bioRxiv

Article Title: A genome language model for mapping DNA replication origins

doi: 10.64898/2026.01.29.702604

Figure Lengend Snippet: a) Schematic representation of ORILINX training on human origins of replication sequences, followed by applying ORILINX predictions in chicken, sheep and mouse without additional adjustments. b) ROC curves showing the ORILINX model performance in 28,490 SNS-seq chicken embryonic fibroblast cell origins of replication and matched number of random non-origin sequences, resulting in an AUC ROC = 0.92 and AUC PR =0.93. c) Same as in b) but for 79,574 SNS-seq sheep primary fibroblast origins and matched number random non-origin sequences from two replicates, resulting in an AUC ROC = 0.93 and AUC PR = 0.94. d) Same as in b) and c) but for publicly available mouse ESC data of 13,004 SNS-seq origins and matched number of random non-origin sequences, resulting in an AUC ROC = 0.81 and AUC PR = 0.85.

Article Snippet: Chicken embryonic fibroblast cells (UMNSAH/DF-1) were purchased from ATCC (ATCC-CRL-3586) and were maintained according to the supplier’s instructions in DMEM (ATCC 30-2002) supplemented with 10% fetal bovine serum (Gibco, 10270-106).

Techniques:

Journal: bioRxiv

Article Title: A genome language model for mapping DNA replication origins

doi: 10.64898/2026.01.29.702604

Figure Lengend Snippet: AUC ROC and AUC PR curves for individual SNS-seq replicates of the sheep fibroblast cells (see Methods). a) Replicate 1 showed an AUC ROC of 0.93 and AUC PR of 0.94. b) Replicate 2 showed an AUC ROC of 0.94 and AUC PR of 0.95.

Article Snippet: Chicken embryonic fibroblast cells (UMNSAH/DF-1) were purchased from ATCC (ATCC-CRL-3586) and were maintained according to the supplier’s instructions in DMEM (ATCC 30-2002) supplemented with 10% fetal bovine serum (Gibco, 10270-106).

Techniques:

Journal: bioRxiv

Article Title: A genome language model for mapping DNA replication origins

doi: 10.64898/2026.01.29.702604

Figure Lengend Snippet: Correlation analyses for a) sheep primary fibroblast, where Spearman’s ρ = 0.81, p-value ≪ 0.0001, and b) mouse embryonic stem cells (mESC), where Spearman’s ρ = 0.78, p-value ≪ 0.0001. For each species, the left panel shows chromosome 1 profiles comparing ORILINX predicted origin probability (red) with origin efficiency inferred from replication timing data (blue), with both signals smoothed using a 10 Mb moving average window. Spearman correlation coefficients are indicated. The right panel shows the corresponding joint density distributions of origin probability and origin efficiency computed genome-wide.

Article Snippet: Chicken embryonic fibroblast cells (UMNSAH/DF-1) were purchased from ATCC (ATCC-CRL-3586) and were maintained according to the supplier’s instructions in DMEM (ATCC 30-2002) supplemented with 10% fetal bovine serum (Gibco, 10270-106).

Techniques: Genome Wide