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
なぜ人はRNA-Seqのリードカウントを負の二項分布に従うと考えるのか
Search
cakkby
April 26, 2022
Science
0
1.4k
なぜ人はRNA-Seqのリードカウントを負の二項分布に従うと考えるのか
なぜ人はRNA-Seqのリードカウントを負の二項分布に従うと考えるのかについて解説してみました
cakkby
April 26, 2022
Tweet
Share
Other Decks in Science
See All in Science
白金鉱業Meetup Vol.16_数理最適化案件のはじめかた・すすめかた
brainpadpr
3
1.8k
baseballrによるMLBデータの抽出と階層ベイズモデルによる打率の推定 / TokyoR118
dropout009
1
440
データベース09: 実体関連モデル上の一貫性制約
trycycle
PRO
0
680
モンテカルロDCF法による事業価値の算出(モンテカルロ法とベイズモデリング) / Business Valuation Using Monte Carlo DCF Method (Monte Carlo Simulation and Bayesian Modeling)
ikuma_w
0
170
Valuable Lessons Learned on Kaggle’s ARC AGI LLM Challenge (PyDataGlobal 2024)
ianozsvald
0
390
眼科AIコンテスト2024_特別賞_6位Solution
pon0matsu
0
400
03_草原和博_広島大学大学院人間社会科学研究科教授_デジタル_シティズンシップシティで_新たな_学び__をつくる.pdf
sip3ristex
0
470
How To Buy, Verified Venmo Accounts in 2025 This year
usaallshop68
2
110
Healthcare Innovation through Business Entrepreneurship
clintwinters
0
230
Masseyのレーティングを用いたフォーミュラレースドライバーの実績評価手法の開発 / Development of a Performance Evaluation Method for Formula Race Drivers Using Massey Ratings
konakalab
0
160
Cross-Media Information Spaces and Architectures (CISA)
signer
PRO
3
31k
データマイニング - グラフデータと経路
trycycle
PRO
1
130
Featured
See All Featured
Keith and Marios Guide to Fast Websites
keithpitt
411
22k
Exploring the Power of Turbo Streams & Action Cable | RailsConf2023
kevinliebholz
34
5.9k
Scaling GitHub
holman
459
140k
Typedesign – Prime Four
hannesfritz
42
2.7k
Code Reviewing Like a Champion
maltzj
524
40k
The Cult of Friendly URLs
andyhume
79
6.5k
Making Projects Easy
brettharned
116
6.3k
Raft: Consensus for Rubyists
vanstee
140
7k
Design and Strategy: How to Deal with People Who Don’t "Get" Design
morganepeng
130
19k
Adopting Sorbet at Scale
ufuk
77
9.4k
Optimizing for Happiness
mojombo
379
70k
Fight the Zombie Pattern Library - RWD Summit 2016
marcelosomers
233
17k
Transcript
None
• • • •
None
None
None
𝑖 𝑙𝑖 𝑒𝑖 𝑁 𝑖 𝑁𝑖 𝑁𝑖 𝑁 = 𝑙𝑖
𝑒𝑖 σ 𝑗 𝑙𝑗 𝑒𝑗 𝑒𝑖 = 1 𝑒𝑖 𝑞𝑖 = 𝑙𝑖 𝑒𝑖 𝑞𝑖
𝑛 𝑔 = σ𝑗 𝑞𝑗 𝑖 𝑞𝑖 𝑁 𝑝 =
𝑞𝑖 /𝑔 𝑖 Pr 𝑁 = 𝑘 = 𝐵𝑖𝑛𝑜𝑚 𝑘|𝑛, 𝑝 = 𝑛 𝑘 𝑝𝑘 1 − 𝑝 𝑛−𝑘
𝑛 𝑔 𝜆 = 𝑛𝑝 = 𝑛𝑞𝑖 𝑔 𝑛, 𝑔
lim 𝜆=𝑛𝑝: fix 𝑛,𝑔→∞ 𝑛 𝑘 𝑝𝑘 1 − 𝑝 𝑛−𝑘 = 𝜆𝑘𝑒−𝜆 𝑘! = 𝑃𝑜𝑖𝑠𝑠𝑜𝑛 𝑘|𝜆 𝜆 = 𝑛𝑝 𝑖 𝜆 𝜆 = 1 𝜆 = 2 𝜆 = 3
𝜆 𝜆 𝜆1 𝜆2
𝜆 𝑖 𝑗 𝜆𝑖𝑗 𝜑 𝜃 𝜃 𝑃 𝑥|𝜃 𝜃
𝑃 𝑥 = 1 𝑀 𝑗=1 𝑀 𝑃 𝑥|𝜃𝑗 ≃ න𝜑 𝜃 𝑃 𝑥|𝜃 𝑑𝜃
𝐺𝑎𝑚𝑚𝑎 𝜆|𝜇, 𝜙 𝑃𝑜𝑖𝑠𝑠𝑜𝑛 𝑥|𝜆 න 0 ∞ 𝐺𝑎𝑚𝑚𝑎 𝜆|𝜙,
𝜇 𝜙 𝑃𝑜𝑖𝑠𝑠𝑜𝑛 𝑥|𝜆 𝑑𝜆 = Γ 𝑥 + 𝜙 Γ 𝑥 + 1 Γ 𝜙 𝜙 𝜇 + 𝜙 𝜙 𝜇 𝜇 + 𝜙 𝑥 = 𝑁𝑒𝑔𝐵𝑖𝑛𝑜𝑚 𝑥|𝜇, 𝜙 𝜇, 𝜙 𝑉 𝑥 = 𝜇 + 𝜇2 𝜙 > 𝜇 𝜙 = ∞
None
𝜙 𝜎2 = 𝜇 𝜎2 = 𝜇 + 𝜇2 𝜙
𝑒𝑖 𝑞𝑖 𝑥𝑖 ′ = 𝑥𝑖 𝑙𝑖 𝐸 𝑥′ =
𝐸 𝑥 𝑙𝑖 = 𝜇 𝑙𝑖 = 𝜇, 𝑉 𝑥′ = 𝑉 𝑥 𝑙𝑖 2 = 𝜇 𝑙𝑖 2 + 𝜇2 𝑙𝑖 2𝜙 = 𝜇 𝑙𝑖 + 𝜇2 𝜙 ≠ 𝜇 + 𝜇2 𝜙
log2 𝑦 = 𝑋𝑓𝑢𝑙𝑙 𝛽 log2 𝑦 = 𝑋𝑟𝑒𝑑𝑢𝑐𝑒𝑑 𝛽
𝑃 𝑌|𝑋𝑓𝑢𝑙𝑙 𝑃 𝑌|𝑋𝑟𝑒𝑑𝑢𝑐𝑒𝑑
None
𝑃 𝑘 = 𝑛 𝑛 − 1 ⋯ 𝑛 −
𝑘 𝑘! 𝑝𝑘 1 − 𝑝 𝑛−𝑘 𝜆 = 𝑛𝑝 𝑛 → ∞ 𝑝 → 0 𝑃 𝑘 ≈ 𝑛𝑘 𝑘! 𝑝𝑘 1 − 𝑝 𝑛 = 𝜆𝑘 𝑘! 1 − 𝜆 𝑛 𝑛 𝑒 1 − 𝜆 𝑛 𝑛 → 𝑒−𝜆 𝑃 𝑘 ≈ 𝜆𝑘 𝑘! 𝑒−𝜆
න 0 ∞ 𝐺𝑎𝑚𝑚𝑎 𝜆|𝜙, 𝜇 𝜙 𝑃𝑜𝑖𝑠𝑠𝑜𝑛 𝑥|𝜆 𝑑𝜆
= න 0 ∞ 𝜆𝜙−1𝑒− 𝜆𝜇 𝜙 𝜙 𝜇 𝜙 Γ 𝜙 𝜆𝑥 𝑥! 𝑒−𝜆𝑑𝜆 = 𝜇 𝜙 𝜙 Γ 𝑥 + 1 Γ 𝜙 න 0 ∞ 𝜆𝜙+𝑥−1𝑒−𝜆 𝜇+𝜙 𝜙 𝑑𝜆 = 𝜙 𝜇 + 𝜙 −𝜙 𝜇 𝜇 + 𝜙 𝜙 Γ 𝑥 + 1 Γ 𝜙 𝜙 𝜇 + 𝜙 𝜙+𝑥 Γ 𝜙 + 𝑥 = Γ 𝜙 + 𝑥 Γ 𝑥 + 1 Γ 𝜙 𝜙 𝜇 + 𝜙 𝑥 𝜇 𝜇 + 𝜙 𝜙 = Γ 𝑥 + 𝜙 Γ 𝑥 + 1 Γ 𝜙 𝜙 𝜇 + 𝜙 𝜙 𝜇 𝜇 + 𝜙 𝑥
• • • • • •