Extended Data Table 1 Robustness of InstructNA for decoding FNA sequences

From: De novo design of functional nucleic acids of aptamers

Models/TFs

Ar

Atf4

Dbp

Egr3

Foxg1

Klf12

Mlx

Nr2e1

Sox10

Srebf1

InstructNA-3mers

0.8759

0.9038

0.7758

0.8765

0.8640

0.9818

0.7180

0.9357

0.8523

0.8263

RaptGen-3mers

0.7952

0.6001

0.5397

0.7259

0.6707

0.7241

0.6105

0.5344

0.7638

0.6564

InstructNA-4mers

0.8413

0.8838

0.8086

0.8480

0.8332

0.9745

0.7406

0.8983

0.7563

0.7714

RaptGen-4mers

0.7186

0.4821

0.5578

0.5636

0.4735

0.5252

0.7139

0.3875

0.6505

0.5198

InstructNA-5mers

0.8034

0.8724

0.8205

0.8075

0.7943

0.9650

0.6931

0.8386

0.6890

0.7185

RaptGen-5mers

0.6529

0.3519

0.5986

0.4357

0.3596

0.3465

0.7809

0.2455

0.5752

0.3970

InstructNA-6mers

0.7622

0.8568

0.8020

0.7598

0.7441

0.9522

0.6390

0.7232

0.6204

0.6548

RaptGen-6mers

0.5868

0.2075

0.6566

0.2796

0.2827

0.2254

0.7553

0.1258

0.5061

0.2918

  1. PCC of k-mer frequencies between real sequences and generated sequences by InstructNA and RaptGen from perturbated representations.
Back to article page