README.md

---
datasets:
- multimolecule/rnacentral
- multimolecule/rfam
- multimolecule/ensembl-genome-browser
- multimolecule/nucleotide
library_name: multimolecule
license: agpl-3.0
mask_token: <mask>
pipeline_tag: fill-mask
tags:
- Biology
- RNA
- ncRNA
- rna
widget:
- example_title: microRNA 21
  mask_index: 11
  mask_index_1based: 12
  masked_char: A
  output:
  - label: U
    score: 0.182346
  - label: A
    score: 0.163327
  - label: G
    score: 0.162945
  - label: X
    score: 0.161102
  - label: C
    score: 0.138805
  pipeline_tag: fill-mask
  sequence_type: ncRNA
  task: fill-mask
  text: UAGCUUAUCAG<mask>CUGAUGUUGA
- example_title: microRNA 146a
  mask_index: 10
  mask_index_1based: 11
  masked_char: A
  output:
  - label: A
    score: 0.174767
  - label: U
    score: 0.173765
  - label: G
    score: 0.160257
  - label: X
    score: 0.158629
  - label: C
    score: 0.130105
  pipeline_tag: fill-mask
  sequence_type: ncRNA
  task: fill-mask
  text: UGAGAACUGA<mask>UUCCAUGGGUU
- example_title: microRNA 155
  mask_index: 15
  mask_index_1based: 16
  masked_char: A
  output:
  - label: U
    score: 0.177485
  - label: G
    score: 0.169577
  - label: A
    score: 0.168204
  - label: X
    score: 0.160879
  - label: C
    score: 0.132323
  pipeline_tag: fill-mask
  sequence_type: ncRNA
  task: fill-mask
  text: UUAAUGCUAAUCGUG<mask>UAGGGGUU
- example_title: RNA component of mitochondrial RNA processing endoribonuclease
  mask_index: 11
  mask_index_1based: 12
  masked_char: A
  output:
  - label: C
    score: 0.226097
  - label: G
    score: 0.184231
  - label: X
    score: 0.160851
  - label: A
    score: 0.138966
  - label: U
    score: 0.115647
  pipeline_tag: fill-mask
  sequence_type: ncRNA
  task: fill-mask
  text: GGUUCGUGCUG<mask>AGGCCUGUAUCCUAGGCUACACACUGAGGACUCUGUUCCUCCCCUUUCCGCCUAGGGGAAAGUCCCCGGACCUCGGGCAGAGAGUGCCACGUGCAUACGCACGUAGACAUUCCCCGCUUCCCACUCCAAAGUCCGCCAAGAAGCGUAUCCCGCUGAGCGGCGUGGCGCGGGGGCGUCAUCCGUCAGCUCCCUCUAGUUACGCAGGCAGUGCGUGUCCGCGCACCAACCACACGGGGCUCAUUCUCAGCGCGGCUGUAAAAAAAAA
- example_title: 7SK small nuclear RNA
  mask_index: 13
  mask_index_1based: 14
  masked_char: A
  output:
  - label: A
    score: 0.968171
  - label: X
    score: 0.006591
  - label: <null>
    score: 0.005137
  - label: '*'
    score: 0.004361
  - label: .
    score: 0.004294
  pipeline_tag: fill-mask
  sequence_type: ncRNA
  task: fill-mask
  text: GGAUGUGAGGGCG<mask>UCUGGCUGCGACAUCUGUCACCCCAUUGAUCGCCAGGGUUGAUUCGGCUGAUCUGGCUGGCUAGGCGGGUGUCCCCUUCCUCCCUCACCGCUCCAUGUGCGUCCCUCCCGAAGCUGCGCGCUCGGUCGAAGAGGACGACCAUCCCCGAUAGAGGAGGACCGGUCUUCGGUCAAGGGUAUACGAGUAGCUGCGCUCCCCUGCUAGAACCUCCAAACAAGCUCUCAAGGUCCAUUUGUAGGAGAACGUAGGGUAGUCAAGCUUCCAAGACUCCAGACACAUCCAAAUGAGGCGCUGCAUGUGGCAGUCUGCCUUUCUUUU
- example_title: telomerase RNA component
  mask_index: 23
  mask_index_1based: 24
  masked_char: A
  output:
  - label: U
    score: 0.191585
  - label: X
    score: 0.154069
  - label: C
    score: 0.152917
  - label: A
    score: 0.144616
  - label: G
    score: 0.132991
  pipeline_tag: fill-mask
  sequence_type: ncRNA
  task: fill-mask
  text: GGGUUGCGGAGGGUGGGCCUGGG<mask>GGGGUGGUGGCCAUUUUUUGUCUAACCCUAACUGAGAAGGGCGUAGGCGCCGUGCUUUUGCUCCCCGCGCGCUGUUUUUCUCGCUGACUUUCAGCGGGCGGAAAAGCCUCGGCCUGCCGCCUUCCACCGUUCAUUCUAGAGCAAACAAAAAAUGUCAGCUGCUGGCCCGUUCGCCCCUCCCGGGGACCUGCGGCGGGUCGCCUGCCCAGCCCCCGAACCCCGCCUGGAGGCCGCGGUCGGCCCGGGGCUUCUCCGGAGGCACCCACUGCCACCGCGAAGAGUUGGGCUCUGUCAGCCGCGGGUCUCUCGGGGGCGAGGGCGAGGUUCAGGCCUUUCAGGCCGCAGGAAGAGGAACGGAGCGAGUCCCCGCGCGCGGCGCGAUUCCCUGAGCUGUGGGACGUGCACCCAGGACUCGGCUCACACAUGC
- example_title: vault RNA 2-1
  mask_index: 12
  mask_index_1based: 13
  masked_char: A
  output:
  - label: G
    score: 0.296311
  - label: A
    score: 0.21848
  - label: X
    score: 0.153834
  - label: U
    score: 0.145476
  - label: C
    score: 0.059464
  pipeline_tag: fill-mask
  sequence_type: ncRNA
  task: fill-mask
  text: CGGGUCGGAGUU<mask>GCUCAAGCGGUUACCUCCUCAUGCCGGACUUUCUAUCUGUCCAUCUCUGUGCUGGGGUUCGAGACCCGCGGGUGCUUACUGACCCUUUUAUGCAA
- example_title: brain cytoplasmic RNA 1
  mask_index: 18
  mask_index_1based: 19
  masked_char: A
  output:
  - label: A
    score: 0.927616
  - label: X
    score: 0.02096
  - label: G
    score: 0.009731
  - label: '|'
    score: 0.007956
  - label: '*'
    score: 0.007946
  pipeline_tag: fill-mask
  sequence_type: ncRNA
  task: fill-mask
  text: GGCCGGGCGCGGUGGCUC<mask>CGCCUGUAAUCCCAGCUCUCAGGGAGGCUAAGAGGCGGGAGGAUAGCUUGAGCCCAGGAGUUCGAGACCUGCCUGGGCAAUAUAGCGAGACCCCGUUCUCCAGAAAAAGGAAAAAAAAAAACAAAAGACAAAAAAAAAAUAAGCGUAACUUCCCUCAAAGCAACAACCCCCCCCCCCCUUU
- example_title: HIV-1 TAR-WT
  mask_index: 13
  mask_index_1based: 14
  masked_char: A
  output:
  - label: G
    score: 0.268135
  - label: U
    score: 0.204114
  - label: A
    score: 0.200433
  - label: X
    score: 0.136009
  - label: '*'
    score: 0.048898
  pipeline_tag: fill-mask
  sequence_type: ncRNA
  task: fill-mask
  text: GGUCUCUCUGGUU<mask>GACCAGAUCUGAGCCUGGGAGCUCUCUGGCUAACUAGGGAACC
- example_title: prion protein (Kanno blood group)
  mask_index: 21
  mask_index_1based: 22
  masked_char: A
  output:
  - label: C
    score: 0.249875
  - label: G
    score: 0.203934
  - label: X
    score: 0.159091
  - label: U
    score: 0.14821
  - label: A
    score: 0.084818
  pipeline_tag: fill-mask
  sequence_type: mRNA
  task: fill-mask
  text: AUGGCGAACCUUGGCUGCUGG<mask>UGCUGGUUCUCUUUGUGGCCACAUGGAGUGACCUGGGCCUCUGC
- example_title: interleukin 10
  mask_index: 11
  mask_index_1based: 12
  masked_char: A
  output:
  - label: A
    score: 0.305148
  - label: U
    score: 0.292251
  - label: X
    score: 0.117275
  - label: '*'
    score: 0.065927
  - label: G
    score: 0.057991
  pipeline_tag: fill-mask
  sequence_type: mRNA
  task: fill-mask
  text: AUGCACAGCUC<mask>GCACUGCUCUGUUGCCUGGUCCUCCUGACUGGGGUGAGGGCC
- example_title: Zaire ebolavirus
  mask_index: 11
  mask_index_1based: 12
  masked_char: A
  output:
  - label: U
    score: 0.294927
  - label: A
    score: 0.286149
  - label: X
    score: 0.127361
  - label: C
    score: 0.094125
  - label: '*'
    score: 0.053796
  pipeline_tag: fill-mask
  sequence_type: mRNA
  task: fill-mask
  text: AAUGUUCAAAC<mask>CUUUGUGAAGCUCUGUUAGCUGAUGGUCUUGCUAAAGCAUUUCCUAGCAAUAUGAUGGUAGUCACAGAGCGUGAGCAAAAAGAAAGCUUAUUGCAUCAAGCAUCAUGGCACCACACAAGUGAUGAUUUUGGUGAGCAUGCCACAGUUAGAGGGAGUAGCUUUGUAACUGAUUUAGAGAAAUACAAUCUUGCAUUUAGAUAUGAGUUUACAGCACCUUUUAUAGAAUAUUGUAACCGUUGCUAUGGUGUUAAGAAUGUUUUUAAUUGGAUGCAUUAUACAAUCCCACAGUGUUAU
- example_title: SARS coronavirus
  mask_index: 14
  mask_index_1based: 15
  masked_char: A
  output:
  - label: U
    score: 0.303839
  - label: X
    score: 0.156264
  - label: C
    score: 0.141866
  - label: A
    score: 0.141266
  - label: G
    score: 0.09792
  pipeline_tag: fill-mask
  sequence_type: mRNA
  task: fill-mask
  text: AUGUUUAUUUUCUU<mask>UUAUUUCUUACUCUCACUAGUGGUAGUGACCUUGACCGGUGCACCACUUUUGAUGAUGUUCAAGCUCCUAAUUACACUCAACAUACUUCAUCUAUGAGGGGGGUUUACUAUCCUGAUGAAAUUUUUAGAUCAGACACUCUUUAUUUAACUCAGGAUUUAUUUCUUCCAUUUUAUUCUAAUGUUACAGGGUUUCAUACUAUUAAUCAUACGUUUGACAACCCUGUCAUACCUUUUAAGGAUGGUAUUUAUUUUGCUGCCACAGAGAAAUCAAAUGUUGUCCGUGGUUGGGUUUUUGGUUCUACCAUGAACAACAAGUCACAGUCGGUGAUUAUUAUUAACAAUUCUACUAAUGUUGUUAUACGAGCAUGUAACUUUGAAUUGUGUGACAACCCUUUCUUUGCUGUUUCUAAACCCAUGGGUACACAGACACAUACUAUGAUAUUCGAUAAUGCAUUUAAAUGCACUUUCGAGUACAUAUCU
- example_title: insulin
  mask_index: 12
  mask_index_1based: 13
  masked_char: A
  output:
  - label: C
    score: 0.596569
  - label: G
    score: 0.135795
  - label: X
    score: 0.094183
  - label: A
    score: 0.04663
  - label: '*'
    score: 0.041054
  pipeline_tag: fill-mask
  sequence_type: mRNA
  task: fill-mask
  text: AUGGCCCUGUGG<mask>UGCGCCUCCUGCCCCUGCUGGCGCUGCUGGCCCUCUGGGGACCUGACCCAGCCGCAGCCUUUGUGAACCAACACCUGUGCGGCUCACACCUGGUGGAAGCUCUCUACCUAGUGUGCGGGGAACGAGGCUUCUUCUACACACCCAAGACCCGCCGGGAGGCAGAGGACCUGCAGGUGGGGCAGGUGGAGCUGGGCGGGGGCCCUGGUGCAGGCAGCCUGCAGCCCUUGGCCCUGGAGGGGUCCCUGCAGAAGCGUGGCAUUGUGGAACAAUGCUGUACCAGCAUCUGCUCCCUCUACCAGCUGGAGAACUACUGCAACUAG
- example_title: cyclin dependent kinase inhibitor 2A
  mask_index: 18
  mask_index_1based: 19
  masked_char: A
  output:
  - label: G
    score: 0.278792
  - label: A
    score: 0.206863
  - label: C
    score: 0.14254
  - label: X
    score: 0.138775
  - label: '*'
    score: 0.070909
  pipeline_tag: fill-mask
  sequence_type: mRNA
  task: fill-mask
  text: AUGGAGCCGGCGGCGGGG<mask>GCAGCAUGGAGCCUUCGGCUGACUGGCUGGCCACGGCCGCGGCCCGGGGUCGGGUAGAGGAGGUGCGGGCGCUGCUGGAGGCGGGGGCGCUGCCCAACGCACCGAAUAGUUACGGUCGGAGGCCGAUCCAGGUCAUGAUGAUGGGCAGCGCCCGAGUGGCGGAGCUGCUGCUGCUCCACGGCGCGGAGCCCAACUGCGCCGACCCCGCCACUCUCACCCGACCCGUGCACGACGCUGCCCGGGAGGGCUUCCUGGACACGCUGGUGGUGCUGCACCGGGCCGGGGCGCGGCUGGACGUGCGCGAUGCCUGGGGCCGUCUGCCCGUGGACCUGGCUGAGGAGCUGGGCCAUCGCGAUGUCGCACGGUACCUGCGCGCGGCUGCGGGGGGCACCAGAGGCAGUAACCAUGCCCGCAUAGAUGCCGCGGAAGGUCCCUCAGACAUCCCCGAUUGA
- example_title: human papillomavirus type 16 E6
  mask_index: 10
  mask_index_1based: 11
  masked_char: A
  output:
  - label: A
    score: 0.23786
  - label: U
    score: 0.202088
  - label: G
    score: 0.150697
  - label: X
    score: 0.13966
  - label: '*'
    score: 0.090803
  pipeline_tag: fill-mask
  sequence_type: mRNA
  task: fill-mask
  text: AUGCACCAAA<mask>GAGAACUGCAAUGUUUCAGGACCCACAGGAGCGACCCAGAAAGUUACCACAGUUAUGCACAGAGCUGCAAACAACUAUACAUGAUAUAAUAUUAGAAUGUGUGUACUGCAAGCAACAGUUACUGCGACGUGAGGUAUAUGACUUUGCUUUUCGGGAUUUAUGCAUAGUAUAUAGAGAUGGGAAUCCAUAUGCUGUAUGUGAUAAAUGUUUAAAGUUUUAUUCUAAAAUUAGUGAGUAUAGACAUUAUUGUUAUAGUUUGUAUGGAACAACAUUAGAACAGCAAUACAACAAACCGUUGUGUGAUUUGUUAAUUAGGUGUAUUAACUGUCAAAAGCCACUGUGUCCUGAAGAAAAGCAAAGACAUCUGGACAAAAAGCAAAGAUUCCAUAAUAUAAGGGGUCGGUGGACCGGUCGAUGUAUGUCUUGUUGCAGAUCAUCAAGAACACGUAGAGAAACCCAGCUGUAA
- example_title: NRAS proto-oncogene
  mask_index: 36
  mask_index_1based: 37
  masked_char: A
  output:
  - label: C
    score: 0.376813
  - label: U
    score: 0.182384
  - label: X
    score: 0.137855
  - label: G
    score: 0.090835
  - label: A
    score: 0.057852
  pipeline_tag: fill-mask
  sequence_type: 5' UTR
  task: fill-mask
  text: GGGGCCGGAAGUGCCGCUCCUUGGUGGGGGCUGUUC<mask>UGGCGGUUCCGGGGUCUCCAACAUUUUUCCCGGCUGUGGUCCUAAAUCUGUCCAAAGCAGAGGCAGUGGAGCUUGAGGUUCUUGCUGGUGUGAA
- example_title: amyloid beta precursor protein
  mask_index: 15
  mask_index_1based: 16
  masked_char: A
  output:
  - label: G
    score: 0.420064
  - label: X
    score: 0.129558
  - label: C
    score: 0.121108
  - label: A
    score: 0.088767
  - label: '*'
    score: 0.063158
  pipeline_tag: fill-mask
  sequence_type: 5' UTR
  task: fill-mask
  text: GUCAGUUUCCUCGGC<mask>GCGGUAGGCGAGAGCACGCGGAGGAGCGUGCGCGGGGGCCCCGGGAGACGGCGGCGGUGGCGGCGCGGGCAGAGCAAGGACGCGGCGGAUCCCACUCGCACAGCAGCGCACUCGGUGCCCCGCGCAGGGUCGCG
- example_title: RUNX family transcription factor 1
  mask_index: 15
  mask_index_1based: 16
  masked_char: A
  output:
  - label: A
    score: 0.240759
  - label: U
    score: 0.191693
  - label: C
    score: 0.188958
  - label: X
    score: 0.148234
  - label: '*'
    score: 0.062019
  pipeline_tag: fill-mask
  sequence_type: 5' UTR
  task: fill-mask
  text: ACUUCUUUGGGCCUC<mask>UAAACAACCACAGAACCACAAGUUGGGUAGCCUGGCAGUGUCAGAAGUCUGAACCCAGCAUAGUGGUCAGCAGGCAGGACGAAUCACACUGAAUGCAAACCACAGGGUUUCGCAGCGUGGUAAAAGAAAUCAUUGAGUCCCCCGCCUUCAGAAGAGGGUGCAUUUUCAGGAGGAAGCG
- example_title: fragile X messenger ribonucleoprotein 1
  mask_index: 15
  mask_index_1based: 16
  masked_char: A
  output:
  - label: C
    score: 0.234132
  - label: G
    score: 0.193929
  - label: X
    score: 0.150817
  - label: A
    score: 0.128924
  - label: U
    score: 0.088382
  pipeline_tag: fill-mask
  sequence_type: 5' UTR
  task: fill-mask
  text: CUCAGUCAGGCGCUC<mask>GCUCCGUUUCGGUUUCACUUCCGGUGGAGGGCCGCCUCUGAGCGGGCGGCGGGCCGACGGCGAGCGCGGGCGGCGGCGGUGACGGAGGCGCCGCUGCCAGGGGGCGUGCGGCAGCGCGGCGGCGGCGGCGGCGGCGGCGGCGGCGGAGGCGGCGGCGGCGGCGGCGGCGGCGGCGGCUGGGCCUCGAGCGCCCGCAGCCCACCUCUCGGGGGCGGGCUCCCGGCGCUAGCAGGGCUGAAGAGAAG
- example_title: MYC proto-oncogene
  mask_index: 10
  mask_index_1based: 11
  masked_char: A
  output:
  - label: U
    score: 0.203058
  - label: G
    score: 0.192648
  - label: X
    score: 0.152914
  - label: C
    score: 0.138098
  - label: A
    score: 0.101209
  pipeline_tag: fill-mask
  sequence_type: 5' UTR
  task: fill-mask
  text: AACUCGCUGU<mask>GUAAUUCCAGCGAGAGGCAGAGGGAGCGAGCGGGCGGCCGGCUAGGGUGGAAGAGCCGGGCGAGCAGAGCUGCGCUGCGGGCGUCCUGGGAAGGGAGAUCCGGAGCGAAUAGGGGGCUUCGCCUCUGGCCCAGCCCUCCCGCUGAUCCCCCAGCCAGCGGUCCGCAACCCUUGCCGCAUCCACGAAACUUUGCCCAUAGCAGCGGGCGGGCACUUUGCACUGGAACUUACAACACCCGAGCAAGGACGCGACUCUCCCGACGCGGGGAGGCUAUUCUGCCCAUUUGGGGACACUUCCCCGCCGCUGCCAGGACCCGCUUCUCUGAAAGGCUCUCCUUGCAGCUGCUUAGACG
- example_title: activating transcription factor 4
  mask_index: 20
  mask_index_1based: 21
  masked_char: A
  output:
  - label: U
    score: 0.209832
  - label: C
    score: 0.178862
  - label: X
    score: 0.15708
  - label: G
    score: 0.136581
  - label: A
    score: 0.118769
  pipeline_tag: fill-mask
  sequence_type: 5' UTR
  task: fill-mask
  text: CAUUUCUACUUUGCCCGCCC<mask>CAGAUGUAGUUUUCUCUGCGCGUGUGCGUUUUCCCUCCUCCCCGCCCUCAGGGUCCACGGCCACCAUGGCGUAUUAGGGGCAGCAGUGCCUGCGGCAGCAUUGGCCUUUGCAGCGGCGGCAGCAGCACCAGGCUCUGCAGCGGCAACCCCCAGCGGCUUAAGCCAUGGCGCUUCUCACGGCAUUCAGCAGCAGCGUUGCUGUAACCGACAAAGACACCUUCGAAUUAAGCACAUUCCUCGAUUCCAGCAAAGCACCGCAAC
- example_title: Human GPI protein p137
  mask_index: 11
  mask_index_1based: 12
  masked_char: A
  output:
  - label: A
    score: 0.291614
  - label: X
    score: 0.155631
  - label: U
    score: 0.129081
  - label: G
    score: 0.125773
  - label: C
    score: 0.123917
  pipeline_tag: fill-mask
  sequence_type: 3' UTR
  task: fill-mask
  text: UUUUUAAAAGG<mask>AAAGAUACCAAAUGCCUGCUGCUACCACCCUUUUCAAUUGCUAUGUUUUGAAAGGCACCAGUAUGUGUUUUAGAUUGAUUUAAAUGUUUCAUUUAAAUCACGGACAGUAGUUUCAGUUCUGAUGGUAUAAGCAAAACAAAUAAAACGUUUAUAAAAGUUGUAUCUUGAAACACUGGUGUUCAACAGCUAGCAGCUUAUGUGAUUCACCCCAUGCCACGUUAGUGUCACAAAUUUUAUGGUUUAUCUCCAGCAACAUUUCUCUAGUACUUGCACUUAUUAUCUGAAUUC
- example_title: nucleophosmin 1
  mask_index: 11
  mask_index_1based: 12
  masked_char: A
  output:
  - label: U
    score: 0.237947
  - label: A
    score: 0.197577
  - label: X
    score: 0.158222
  - label: G
    score: 0.119338
  - label: C
    score: 0.111704
  pipeline_tag: fill-mask
  sequence_type: 3' UTR
  task: fill-mask
  text: GAAAAUAGUUU<mask>AACAAUUUGUUAAAAAAUUUUCCGUCUUAUUUCAUUUCUGUAACAGUUGAUAUCUGGCUGUCCUUUUUAUAAUGCAGAGUGAGAACUUUCCCUACCGUGUUUGAUAAAUGUUGUCCAGGUUCUAUUGCCAAGAAUGUGUUGUCCAAAAUGCCUGUUUAGUUUUUAAAGAUGGAACUCCACCCUUUGCUUGGUUUUAAGUAUGUAUGGAAUGUUAUGAUAGGACAUAGUAGUAGCGGUGGUCAGACAUGGAAAUGGUGGGGAGACAAAAAUAUACAUGUGAAAUAAAACUCAGUAUUUUAAUAAAGUAGCACGGUUUCUAUUGA
- example_title: superoxide dismutase 1
  mask_index: 12
  mask_index_1based: 13
  masked_char: A
  output:
  - label: C
    score: 0.226842
  - label: A
    score: 0.170614
  - label: X
    score: 0.15816
  - label: U
    score: 0.149522
  - label: G
    score: 0.108129
  pipeline_tag: fill-mask
  sequence_type: 3' UTR
  task: fill-mask
  text: ACAUUCCCUUGG<mask>UGUAGUCUGAGGCCCCUUAACUCAUCUGUUAUCCUGCUAGCUGUAGAAAUGUAUCCUGAUAAACAUUAAACACUGUAAUCUUAAAAGUGUAAUUGUGUGACUUUUUCAGAGUUGCUUUAAAGUACCUGUAGUGAGAAACUGAUUUAUGAUCACUUGGAAGAUUUGUAUAGUUUUAUAAAACUCAGUUAAAAUGUCUGUUUCAAUGACCUGUAUUUUGCCAGACUUAAAUCACAGAUGGGUAUUAAACUUGUCAGAAUUUCUUUGUCAUUCAAGCCUGUGAAUAAAAACCCUGUAUGGCACUUAUUAUGAGGCUAUUAAAAGAAUCCAAAUUCAAACUAAA
- example_title: hemoglobin subunit alpha 2
  mask_index: 13
  mask_index_1based: 14
  masked_char: A
  output:
  - label: G
    score: 0.368338
  - label: C
    score: 0.150316
  - label: X
    score: 0.1398
  - label: U
    score: 0.118529
  - label: A
    score: 0.058205
  pipeline_tag: fill-mask
  sequence_type: 3' UTR
  task: fill-mask
  text: CUGGAGCCUCGGU<mask>GCCGUUCCUCCUGCCCGCUGGGCCUCCCAACGGGCCCUCCUCCCCUCCUUGCACCGGCCCUUCCUGGUCUUUGAAUAAAGUCUGAGUGGGCAGCA
- example_title: BRAF proto-oncogene
  mask_index: 12
  mask_index_1based: 13
  masked_char: A
  output:
  - label: A
    score: 0.365678
  - label: U
    score: 0.186248
  - label: G
    score: 0.157475
  - label: X
    score: 0.115548
  - label: '*'
    score: 0.059832
  pipeline_tag: fill-mask
  sequence_type: 3' UTR
  task: fill-mask
  text: AACAAAUGAGUG<mask>GAGAGUUCAGGAGAGUAGCAACAAAAGGAAAAUAAAUGAACAUAUGUUUGCUUAUAUGUUAAAUUGAAUAAAAUACUCUCUUUUUUUUUAAGGUGAACCAAAGAACACUUGUGUGGUUAAAGACUAGAUAUAAUUUUUCCCCAAACUAAAAUUUAUACUUAACAUUGGAUUUUUAACAUCCAAGGGUUAAAAUACAUAGACAUUGCUAAAAAUUGGCAGAGCCUCUUCUAGAGGCUUUACUUUCUGUUCCGGGUUUGUAUCAUUCACUUGGUUAUUUUAAGUAGUAAACUUCAGUUUCUCAUGCAACUUUUGUUGCCAGCUAUCACAUGUCCACUAGGGACUCCAGAAGAAGACCCUACCUAUGCCUGUGUUUGCAGGUGAGAAGUUGGCAGUCGGUUAGCCUGGG
- example_title: H3 clustered histone 1
  mask_index: 17
  mask_index_1based: 18
  masked_char: A
  output:
  - label: G
    score: 0.24184
  - label: A
    score: 0.191124
  - label: X
    score: 0.162343
  - label: C
    score: 0.129291
  - label: U
    score: 0.11623
  pipeline_tag: fill-mask
  sequence_type: 3' UTR
  task: fill-mask
  text: UUACUGUGGUCUCUCUG<mask>CGGUCCAAGCAAAGGCUCUUUUCAGAGCCACCACCUUUUC
---

# RiNALMo

Pre-trained model on non-coding RNA (ncRNA) using a masked language modeling (MLM) objective.

## Disclaimer

This is an UNOFFICIAL implementation of the [RiNALMo: General-Purpose RNA Language Models Can Generalize Well on Structure Prediction Tasks](https://doi.org/10.48550/arXiv.2403.00043) by Rafael Josip Penić, et al.

The OFFICIAL repository of RiNALMo is at [lbcb-sci/RiNALMo](https://github.com/lbcb-sci/RiNALMo).

> [!TIP]
> The MultiMolecule team has confirmed that the provided model and checkpoints are producing the same intermediate representations as the original implementation.

**The team releasing RiNALMo did not write this model card for this model so this model card has been written by the MultiMolecule team.**

## Model Details

RiNALMo is a [bert](https://huggingface.co/google-bert/bert-base-uncased)-style model pre-trained on a large corpus of non-coding RNA sequences in a self-supervised fashion. This means that the model was trained on the raw nucleotides of RNA sequences only, with an automatic process to generate inputs and labels from those texts. Please refer to the [Training Details](#training-details) section for more information on the training process.

### Variants

- **[multimolecule/rinalmo-giga](https://huggingface.co/multimolecule/rinalmo-giga)**: The RiNALMo model with 650 million parameters.
- **[multimolecule/rinalmo-mega](https://huggingface.co/multimolecule/rinalmo-mega)**: The RiNALMo model with 150 million parameters.
- **[multimolecule/rinalmo-micro](https://huggingface.co/multimolecule/rinalmo-micro)**: The RiNALMo model with 30 million parameters.

### Model Specification

<table>
<thead>
  <tr>
    <th>Variants</th>
    <th>Num Layers</th>
    <th>Hidden Size</th>
    <th>Num Heads</th>
    <th>Intermediate Size</th>
    <th>Num Parameters (M)</th>
    <th>FLOPs (G)</th>
    <th>MACs (G)</th>
    <th>Max Num Tokens</th>
  </tr>
</thead>
<tbody>
  <tr>
    <td>RiNALMo-Giga</td>
    <td>33</td>
    <td>1280</td>
    <td rowspan="3">20</td>
    <td>5120</td>
    <td>650.88</td>
    <td>709.59</td>
    <td>354.31</td>
    <td rowspan="3">1022</td>
  </tr>
  <tr>
    <td>RiNALMo-Mega</td>
    <td>30</td>
    <td>640</td>
    <td>2560</td>
    <td>148.04</td>
    <td>171.76</td>
    <td>85.54</td>
  </tr>
  <tr>
    <td><b>RiNALMo-Micro</b></td>
    <td>12</td>
    <td>480</td>
    <td>1920</td>
    <td>33.48</td>
    <td>40.26</td>
    <td>20.01</td>
  </tr>
</tbody>
</table>

### Links

- **Code**: [multimolecule.rinalmo](https://github.com/DLS5-Omics/multimolecule/tree/master/multimolecule/models/rinalmo)
- **Data**: [multimolecule/rnacentral](https://huggingface.co/datasets/multimolecule/rnacentral)
- **Paper**: [RiNALMo: General-Purpose RNA Language Models Can Generalize Well on Structure Prediction Tasks](https://doi.org/10.48550/arXiv.2403.00043)
- **Developed by**: Rafael Josip Penić, Tin Vlašić, Roland G. Huber, Yue Wan, Mile Šikić
- **Model type**: [BERT](https://huggingface.co/google-bert/bert-base-uncased)
- **Original Repository**: [lbcb-sci/RiNALMo](https://github.com/lbcb-sci/RiNALMo)

## Usage

The model file depends on the [`multimolecule`](https://multimolecule.danling.org) library. You can install it using pip:

```bash
pip install multimolecule
```

### Direct Use

#### Masked Language Modeling

You can use this model directly with a pipeline for masked language modeling:

```python
import multimolecule  # you must import multimolecule to register models
from transformers import pipeline

predictor = pipeline("fill-mask", model="multimolecule/rinalmo-giga")
output = predictor("gguc<mask>cucugguuagaccagaucugagccu")
```

### Downstream Use

#### Extract Features

Here is how to use this model to get the features of a given sequence in PyTorch:

```python
from multimolecule import RnaTokenizer, RiNALMoModel


tokenizer = RnaTokenizer.from_pretrained("multimolecule/rinalmo-giga")
model = RiNALMoModel.from_pretrained("multimolecule/rinalmo-giga")

text = "UAGCUUAUCAGACUGAUGUUG"
input = tokenizer(text, return_tensors="pt")

output = model(**input)
```

#### Sequence Classification / Regression

> [!NOTE]
> This model is not fine-tuned for any specific task. You will need to fine-tune the model on a downstream task to use it for sequence classification or regression.

Here is how to use this model as backbone to fine-tune for a sequence-level task in PyTorch:

```python
import torch
from multimolecule import RnaTokenizer, RiNALMoForSequencePrediction


tokenizer = RnaTokenizer.from_pretrained("multimolecule/rinalmo-giga")
model = RiNALMoForSequencePrediction.from_pretrained("multimolecule/rinalmo-giga")

text = "UAGCUUAUCAGACUGAUGUUG"
input = tokenizer(text, return_tensors="pt")
label = torch.tensor([1])

output = model(**input, labels=label)
```

#### Token Classification / Regression

> [!NOTE]
> This model is not fine-tuned for any specific task. You will need to fine-tune the model on a downstream task to use it for token classification or regression.

Here is how to use this model as backbone to fine-tune for a nucleotide-level task in PyTorch:

```python
import torch
from multimolecule import RnaTokenizer, RiNALMoForTokenPrediction


tokenizer = RnaTokenizer.from_pretrained("multimolecule/rinalmo-giga")
model = RiNALMoForTokenPrediction.from_pretrained("multimolecule/rinalmo-giga")

text = "UAGCUUAUCAGACUGAUGUUG"
input = tokenizer(text, return_tensors="pt")
label = torch.randint(2, (len(text), ))

output = model(**input, labels=label)
```

#### Contact Classification / Regression

> [!NOTE]
> This model is not fine-tuned for any specific task. You will need to fine-tune the model on a downstream task to use it for contact classification or regression.

Here is how to use this model as backbone to fine-tune for a contact-level task in PyTorch:

```python
import torch
from multimolecule import RnaTokenizer, RiNALMoForContactPrediction


tokenizer = RnaTokenizer.from_pretrained("multimolecule/rinalmo-giga")
model = RiNALMoForContactPrediction.from_pretrained("multimolecule/rinalmo-giga")

text = "UAGCUUAUCAGACUGAUGUUG"
input = tokenizer(text, return_tensors="pt")
label = torch.randint(2, (len(text), len(text)))

output = model(**input, labels=label)
```

## Training Details

RiNALMo used Masked Language Modeling (MLM) as the pre-training objective: taking a sequence, the model randomly masks 15% of the tokens in the input then runs the entire masked sentence through the model and has to predict the masked tokens. This is comparable to the Cloze task in language modeling.

### Training Data

The RiNALMo model was pre-trained on a cocktail of databases including [RNAcentral](https://rnacentral.org), [Rfam](https://rfam.org), [Ensembl Genome Browser](https://ensembl.org), and [Nucleotide](https://ncbi.nlm.nih.gov/nucleotide).
The training data contains 36 million unique ncRNA sequences.

To ensure sequence diversity in each training batch, RiNALMo clustered the sequences with [MMSeqs2](https://github.com/soedinglab/MMseqs2) into 17 million clusters and then sampled each sequence in the batch from a different cluster.

RiNALMo preprocessed all tokens by replacing "U"s with "T"s.

Note that during model conversions, "T" is replaced with "U". [`RnaTokenizer`][multimolecule.RnaTokenizer] will convert "T"s to "U"s for you, you may disable this behaviour by passing `replace_T_with_U=False`.

### Training Procedure

#### Preprocessing

RiNALMo used masked language modeling (MLM) as the pre-training objective. The masking procedure is similar to the one used in BERT:

- Mask rate: 15%
- Replacement: `<mask>` for 80% of masked tokens
- Replacement: random token for 10% of masked tokens
- Replacement: unchanged token for 10% of masked tokens

#### Pre-training

The model was trained on 7 NVIDIA A100 GPUs with 80GiB memories.

- Batch Size: 1344
- Epochs: 6
- Learning rate: 5e-5
- Learning rate scheduler: Cosine
- Learning rate warm-up: 2,000 steps
- Learning rate minimum: 1e-5
- Dropout: 0.1

## Citation

```bibtex
@ARTICLE{Penic2025-qf,
  title     = "{RiNALMo}: general-purpose {RNA} language models can generalize
               well on structure prediction tasks",
  author    = "Peni{\'c}, Rafael Josip and Vla{\v s}i{\'c}, Tin and Huber,
               Roland G and Wan, Yue and {\v S}iki{\'c}, Mile",
  abstract  = "While RNA has recently been recognized as an interesting
               small-molecule drug target, many challenges remain to be
               addressed before we take full advantage of it. This emphasizes
               the necessity to improve our understanding of its structures and
               functions. Over the years, sequencing technologies have produced
               an enormous amount of unlabeled RNA data, which hides a huge
               potential. Motivated by the successes of protein language
               models, we introduce RiboNucleic Acid Language Model (RiNALMo)
               to unveil the hidden code of RNA. RiNALMo is the largest RNA
               language model to date, with 650M parameters pre-trained on 36M
               non-coding RNA sequences from several databases. It can extract
               hidden knowledge and capture the underlying structure
               information implicitly embedded within the RNA sequences.
               RiNALMo achieves state-of-the-art results on several downstream
               tasks. Notably, we show that its generalization capabilities
               overcome the inability of other deep learning methods for
               secondary structure prediction to generalize on unseen RNA
               families.",
  journal   = "Nature Communications",
  publisher = "Springer Science and Business Media LLC",
  volume    =  16,
  number    =  1,
  pages     = "5671",
  month     =  jul,
  year      =  2025,
  copyright = "https://creativecommons.org/licenses/by-nc-nd/4.0",
  language  = "en"
}
```

> [!NOTE]
> The artifacts distributed in this repository are part of the MultiMolecule project.
> If MultiMolecule supports your research, please cite the MultiMolecule project as follows:

```bibtex
@software{chen_2024_12638419,
  author    = {Chen, Zhiyuan and Zhu, Sophia Y.},
  title     = {MultiMolecule},
  doi       = {10.5281/zenodo.12638419},
  publisher = {Zenodo},
  url       = {https://doi.org/10.5281/zenodo.12638419},
  year      = 2024,
  month     = may,
  day       = 4
}
```

## Contact

Please use GitHub issues of [MultiMolecule](https://github.com/DLS5-Omics/multimolecule/issues) for any questions or comments on the model card.

Please contact the authors of the [RiNALMo paper](https://doi.org/10.48550/arXiv.2403.00043) for questions or comments on the paper/model.

## License

This model implementation is licensed under the [GNU Affero General Public License](license.md).

For additional terms and clarifications, please refer to our [License FAQ](license-faq.md).

```spdx
SPDX-License-Identifier: AGPL-3.0-or-later
```