Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
aliFreeFold: an alignment-free approach to predict secondary structure from homologous RNA sequences
Search
Yuki_Hamaguchi
August 18, 2018
Science
0
82
aliFreeFold: an alignment-free approach to predict secondary structure from homologous RNA sequences
ISMB読み会2018で紹介したaliFreeFoldに関する発表スライドです。
Yuki_Hamaguchi
August 18, 2018
Tweet
Share
Other Decks in Science
See All in Science
FIBA W杯の日本代表って組み合わせ次第で2次ラウンド行けたんじゃね?をデータで検証
saltcooky12
0
200
Pandas 2 vs Polars vs Dask (PyDataGlobal 2023 December)
ianozsvald
0
360
Machine Learning for Materials (Lecture 4)
aronwalsh
0
650
Direct Preference Optimization
zchenry
0
110
Running llama.cpp on the CPU
ianozsvald
0
190
Scoring probability model based on service landing location and ranking points in men’s professional tennis matches
konakalab
0
140
外国人が思う日本語のむずかしいところ
jamashita
0
330
ABEMAの効果検証事例〜効果の異質性を考える〜
s1ok69oo
3
1.4k
「国と音楽」 ~spotifyrを用いて~ #muana
bob3bob3
2
310
Microbiology Labs.
maleehafatima
0
120
Cross-Media Information Spaces and Architectures (CISA)
signer
PRO
3
24k
Machine Learning for Materials (Lecture 9)
aronwalsh
0
120
Featured
See All Featured
Easily Structure & Communicate Ideas using Wireframe
afnizarnur
185
15k
Designing Dashboards & Data Visualisations in Web Apps
destraynor
225
51k
Adopting Sorbet at Scale
ufuk
66
8.5k
Six Lessons from altMBA
skipperchong
19
2.9k
Raft: Consensus for Rubyists
vanstee
130
6.2k
Build your cross-platform service in a week with App Engine
jlugia
223
17k
Product Roadmaps are Hard
iamctodd
43
9.6k
Put a Button on it: Removing Barriers to Going Fast.
kastner
58
3k
Teambox: Starting and Learning
jrom
126
8.4k
How to Create Impact in a Changing Tech Landscape [PerfNow 2023]
tammyeverts
12
1.4k
Designing with Data
zakiwarfel
94
4.8k
The Language of Interfaces
destraynor
150
22k
Transcript
BMJ'SFF'PME dΞϥΠϝϯτϑϦʔͳ 3/"ೋ࣍ߏ༧ଌख๏d հऀɿޱ༔و ౦ژେֶେֶӃ৽ྖҬՊֶݚڀՊ ϝσΟΧϧใੜ໋ઐ߈ ؠ࡚ݚڀࣨ M2 *4.#+PVSOBM
ࣗݾհ ۙͰࢀՃ͢Δֶձɾݚڀձ ɹɾຊਐԽֶձ ɹ ɹɾੜ໋ͷىݯٴͼਐԽֶձՆͷֶߍ ڵຯ͋Δ ɹਅ֩ੜͷىݯ ಛʹϛτίϯυϦΞ ຊਐԽֶձཁࢫू1 ̍
3/"ೋ࣍ߏ༧ଌͷΞϓϩʔν ᶃҰͭͷ3/"Ԙجྻ͔Β༧ଌ ᶄ૬ಉ3/"Ԙجྻ͔Β༧ଌ ࠷খࣗ༝ΤωϧΪʔߏମ .'& Λࢉग़ ίϯηϯαεͳೋ࣍ߏΛࢉग़ ૯߹ԘجରܗҐஔڑ ࣗ
༝ Τ ω ϧ Ϊ | Methods 103 (2016) 86–98 ૬ಉ3/"ͦΕͧΕΛൺֱ͠ɺ ೋ࣍ߏͷอଘੑΛݕग़ ࣗ༝ΤωϧΪʔύϥϝʔλʹج͖ͮɺ ೋ࣍ߏީิΛཏతʹ୳ࡧɾධՁ ̎ MJGFTDJFODFECKQIPVLPLVQEG#@pOBMQEG BMJ'SFF'PME
ᶃ3/"४࠷దߏΛ༻͍ͯ ɹҎ্ͷਖ਼ղԘجର ɹΛ࣋ͭදతͳ3/"ߏ ɹΛग़ྗ͢Δ ࣗ ༝ Τ ω ϧ Ϊ
| ૯߹ԘجରܗҐஔڑ BMJ'SFF'PMEͷڧΈ ̏ 41"34& 5VSCP'PMEJJ ᶄ࣌ؒܭࢉྔ͕͍ ɹΞϥΠϝϯτϑϦʔख๏ ɹैདྷΑΓɾਫ਼ ɹͷόϥϯε͕ྑ͍ 3/"४࠷ద TVCPQUJNBM ߏ ʼ IUUQTPQFOXFUXBSFPSHXJLJ 8JLJPNJDT3/"@TFDPOEBSZ@TUSVDUVSF@QSFEJDUJPO .94$"3/" Ex) Saccharomyces cerevisiaeͷU3/" 1IF
BMJ'SFF'PMEͷΞϓϩʔνख๏ᶃ input: ૬ಉRNAྻ ̐ :@ "$$6"$66("$6$"($((66"("(6"6 $($666$"6"$(($("("(6$"66((6 6$"""6$$""6"(6"((6" "'@ (((($$66"($6$"($6((("("($($
$6($666($"$($"(("((6$"($((6 6$("6$$$($6"(($6$$" "+@ 6$$6$"(6"($6$"(6((6"("($((6 $(($6(66""$$("66((6$(6"((66 $(""6$$6"$66(((("( 9@ "$$6"$66""$6$"(6((66"("(6"$ 6($666$"6"$(($((("(($"66((6 6$"""6$$""6"(6"((6" 7@ "$6666"""(("6"(6"(666"6$$(6 6((6$66"((""$$""""""66((6($ ""$6$$"""6""""(6" ᶃ֤૬ಉ3/"ྻͷ४࠷దߏΛ3/"TVCPQU Λ༻͍ͯࢉग़ ᶄۙྡϞνʔϑߏ ʢϚϧνϧʔϓɾεςϜʣͷΛࢉग़ ۙྡϞνʔϑߏଞʹϔΞϐϯϧʔϓɺΠϯλʔφϧϧʔϓɺ ɹɹɹɹɹɹɹɹɹόϧδɺΤΫελʔφϧϧʔϓ `ɺ` ɺγϡʔυίϯτ 7JFOOB3/"ύοέʔδͷҰͭ
BMJ'SFF'PMEͷΞϓϩʔνख๏ᶄ output දߏ ̑ ᶅϞνʔϑ͝ͱʹอଘࢦΛܭࢉɹ Τϯτϩϐʔܭࢉ ᶆ֤ϞνʔϑͷอଘࢦʹॏΈ͚ͮ ʢอଘࢦʹ࠷େείΞΛ͔͚Δʣ ᶇ֤Ϟνʔϑͷอଘࢦ͔Βશ૬ಉ3/"ͷॏ৺తߏΛࢉग़
ᶈ֤૬ಉ3/"ͷதͰॏ৺తߏʹ͍ۙͷΛநग़
S3/" U3/" άϧʔϓᶘΠϯτϩϯ 6TO3/" 3/BTF1 UN3/" 4&$*4ʜ 5SFOET#JPDIFN4DJ QQr
IUUQSOBVDTDFEVSOBDFOUFS SJCPTPNF@JNBHFTIUNM *6#.#-JGF Џ༧ଌߏͰ߹கͨ͠Ԙجର ൺֱ͢Δ༧ଌπʔϧɹ(7ͭ) 3/"DBTU طଘͷΞϥΠϝϯτϑϦʔख๏ 3/"TQB .94$"3/" $FOUSPJE'PME 3/"BMJGPME 41"34& 5VSCP'PMEJJ Dataset: طͷ૬ಉRNAྻ σʔληοτͱධՁํ๏ ධՁํ๏: Sensivity, PPV, MCC ਅͷ݁Ռ ରܗ Ԙج ରඇܗ Ԙج ରܗ Ԙج 51 '1 ରඇܗ Ԙج '/ 5/ ༧ ଌ ݁ Ռ ✖51 '/ ˓5/ '/
ೋ࣍ߏ༧ଌͷਖ਼ղԘجର ̓ ฏ ۉ ਖ਼ ղ ׂ ߹ (%)
ݸ M F E ݸ ݸ c ݸ ̍ݸ ४࠷దߏσʔλ .'&ߏΑΓ४࠷దߏͷํ͕ԘجରΛਖ਼͘͠༧ଌͰ͖Δ ४࠷దߏσʔλ͕૿Ճ͢Δ΄Ͳ༧ଌੑೳ্͕ঢ͢Δ
3/"ೋ࣍ߏ༧ଌੑೳᶃ ̔ ᶃ"MJ'SFF'PME3/"DBTUΑΓ.$$ 117 4&/4શͯͰ༏Ε͍ͯΔ ᶄ"MJ'SFF'PMEͱ5VSCP'PMEᶘ͕࠷ྑͷ༧ଌπʔϧͰ͋Δ ධՁࢦඪ ରσʔληοτ OD3/"GBNJMZ
̕ ૬ಉ3/" "MJ'SFF'PMEσʔληοτΛ૿Ճͤ͞Δͱ ͞Βʹ༧ଌೳྗ্͕ঢ͢Δ 3/"ೋ࣍ߏ༧ଌੑೳᶄ
3/"ೋ࣍ߏ༧ଌޮ ฏۉԘج OU ࣌ ؒ (log) ୯Ґඵ "MJ'SFF'PME 3/"BMJGPME
$FOUSPJE'PME 5VSCPGPMEJJ "MJ'SFF'PME$FOUSPJE'PME5VSCPGPMEͱ ಉͷ͍ղੳΛ༗͍ͯ͠Δ 3/"BMJGPMEʹରͯ͠ ྼ͍ͬͯΔ ➕ *ཧ࣌ؒܭࢉྔ O(al2+mn),O(al4+mn) B૬ಉԘجର NϞνʔϑɺOྻ O(l2),O(l4)3/"TVCPQU
ࠓޙͷओͳվળ ᶃ४࠷దߏͷܭࢉπʔϧมߋ ɹ 3/"TVCPQU ˠWT@TVCPQU γϡʔυϊοτߏ ͷݕग़ਫ਼Λ্͛Δ
3/BTF1ͷ༧ଌਫ਼Λ ͋͛ΒΕΔ ᶄอଘࢦͷΑΓਖ਼֬ͳܭࢉ 3/" r ྗֶత౷ܭख๏ͷ׆༻ ଟมྔղੳ ˠΑΓదͳදߏͷݕग़ ࠷ ྑ ४ ࠷ ద ߏ ͔ Β ͷ Ԙ ج ର ڑ .'& ͔ΒͷԘجରڑ γϡʔυϊοτߏ .PM#JP4ZTU
*RNAfold (web)Λͬͯඳը ࣮ࡍʹBMJ'SFF'PMEΛΒͤͯΈͨ ϚπϤΠάαͷ৭ૉମU3/""TO DGU3/"TDBO4& ͰͷU3/"ߏ༧ଌ αϯϓϧσʔλ U3/"ͷೋ࣍ߏ༧ଌैདྷ๏ΑΓྑͦ͞͏ʹݟ͑Δ ΞϯνίυϯྖҬ
DG$FOUSPJE'PME ͰͷU3/"ߏ༧ଌ
ίϝϯτ Cell, Vol. 113, 383–394, May 2, 2003, 7MPPQͷΑ͏ͳߏϞνʔϑΛ
ଞͷU3/"σʔλͰڞ༗͍͕ͯͨ͠ɺ ग़ྗ͞ΕͨදߏͷΈ ͦͷϞνʔϑΛอ༗͍ͯ͠ͳ͔ͬͨ Մೳੑ͕͋ΔͷͰʁ