Human genome promoter tissue-invariant test chromosomes 20 and 21 (SegmentNT human genome annotation split)

BPNet arch. large on SegmentNT human genome annotation promoter tissue-invariant auPRC: promoter tissue-invariant: per-nucleotide annotation (auPRC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

BPNet arch. large on SegmentNT human genome annotation promoter tissue-invariant MCC: promoter tissue-invariant: per-nucleotide annotation (MCC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

BPNet arch. on SegmentNT human genome annotation promoter tissue-invariant auPRC: promoter tissue-invariant: per-nucleotide annotation (auPRC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

BPNet arch. on SegmentNT human genome annotation promoter tissue-invariant MCC: promoter tissue-invariant: per-nucleotide annotation (MCC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

Random-Init (only head) on SegmentNT human genome annotation promoter tissue-invariant auPRC: promoter tissue-invariant: per-nucleotide annotation (auPRC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

Random-Init (only head) on SegmentNT human genome annotation promoter tissue-invariant MCC: promoter tissue-invariant: per-nucleotide annotation (MCC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

Random-Init on SegmentNT human genome annotation promoter tissue-invariant auPRC: promoter tissue-invariant: per-nucleotide annotation (auPRC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

Random-Init on SegmentNT human genome annotation promoter tissue-invariant MCC: promoter tissue-invariant: per-nucleotide annotation (MCC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentBorzoi-30kb on SegmentNT human genome annotation promoter tissue-invariant auPRC: promoter tissue-invariant: per-nucleotide annotation (auPRC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentBorzoi-30kb on SegmentNT human genome annotation promoter tissue-invariant MCC: promoter tissue-invariant: per-nucleotide annotation (MCC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentBorzoi-524kb on SegmentNT human genome annotation promoter tissue-invariant auPRC: promoter tissue-invariant: per-nucleotide annotation (auPRC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentBorzoi-524kb on SegmentNT human genome annotation promoter tissue-invariant MCC: promoter tissue-invariant: per-nucleotide annotation (MCC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentEnformer-196kb on SegmentNT human genome annotation promoter tissue-invariant auPRC: promoter tissue-invariant: per-nucleotide annotation (auPRC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentEnformer-196kb on SegmentNT human genome annotation promoter tissue-invariant MCC: promoter tissue-invariant: per-nucleotide annotation (MCC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentEnformer-30kb on SegmentNT human genome annotation promoter tissue-invariant auPRC: promoter tissue-invariant: per-nucleotide annotation (auPRC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentEnformer-30kb on SegmentNT human genome annotation promoter tissue-invariant MCC: promoter tissue-invariant: per-nucleotide annotation (MCC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentNT-10kb on SegmentNT human genome annotation promoter tissue-invariant auPRC: promoter tissue-invariant: per-nucleotide annotation (auPRC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentNT-10kb on SegmentNT human genome annotation promoter tissue-invariant MCC: promoter tissue-invariant: per-nucleotide annotation (MCC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentNT-20kb on SegmentNT human genome annotation promoter tissue-invariant auPRC: promoter tissue-invariant: per-nucleotide annotation (auPRC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentNT-20kb on SegmentNT human genome annotation promoter tissue-invariant MCC: promoter tissue-invariant: per-nucleotide annotation (MCC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentNT-30kb on SegmentNT human genome annotation promoter tissue-invariant auPRC: promoter tissue-invariant: per-nucleotide annotation (auPRC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentNT-30kb on SegmentNT human genome annotation promoter tissue-invariant MCC: promoter tissue-invariant: per-nucleotide annotation (MCC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentNT-3kb (NTv1 human; 2.5B) on SegmentNT human genome annotation promoter tissue-invariant auPRC: promoter tissue-invariant: per-nucleotide annotation (auPRC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentNT-3kb (NTv1 human; 2.5B) on SegmentNT human genome annotation promoter tissue-invariant MCC: promoter tissue-invariant: per-nucleotide annotation (MCC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported

SegmentNT-3kb (only head) on SegmentNT human genome annotation promoter tissue-invariant auPRC: promoter tissue-invariant: per-nucleotide annotation (auPRC)

Test chromosomes 20 and 21; validation chromosome 22; training remaining chromosomes. Test chunks with genes homologous to train/validation genes excluded using Ensembl BioMart accessed 2024-05-08; homologous distal regulatory elements not excluded. Ten test-set samplings with sliding windows beginning at different genomic starting positions. Mean plus reported standard deviation; not ten training seeds. Best validation checkpoint by average MCC across 14 elements. Per-nucleotide predictions pooled across test sequences separately for each genomic element. MCC and auPRC are separate metrics, not cross-element or cross-table pooled comparisons.

Aggregation: Not reported