超算上百億年找不到的答案，百度這個算法十分鐘內找到了最優解

{"type":"doc","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"blockquote","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"size","attrs":{"size":10}}],"text":"作者 | 百度研究院"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"size","attrs":{"size":10}}],"text":"編輯 | 劉燕"}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"疫情期間，mRNA疫苗以更高的安全性和有效性，更快的研發和生產速度被寄予厚望，成爲疫情防控的重要武器，也是各國生命科學領域的關鍵競爭點。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"雖然目前mRNA技術已被成功商業化，但全球的科研團隊仍面臨一個統一難題：mRNA疫苗普遍穩定性低，致使它非常“脆弱”，很容易因爲降解而失效，這成爲mRNA疫苗研發過程中的巨大阻礙。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"mRNA的這種不穩定性給mRNA疫苗的存儲、運輸和免疫原性都帶來了挑戰，這些問題的解決與mRNA疫苗序列的設計息息相關。目前業內普遍認可的破解之道，就是設計出具有穩定結構且蛋白質翻譯效率高的mRNA序列。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"雖然找到了解決路徑，但是要想破局並不容易。序列設計所需要的龐大的計算量，成爲研發過程中最嚴峻的“攔路虎”。與新冠病毒刺突蛋白對應的mRNA，若採用最傳統的遍歷方法去尋找一條穩定的序列，需要查看10^632次方個mRNA序列。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"打個比方，假設一臺超級計算機每秒計算一個序列，在宇宙誕生到現在的138億年時間裏，可能連潛在序列億萬分之一都無法搜索完成。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"面對如此龐大的搜索空間，必須要有全新的算法，才能在有限的時間內尋找到那條最穩定有效的疫苗序列。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"早在2018年，百度研究院就開展了 RNA 二級結構領域的研究。得益於這幾年在生物計算領域的研究積累，在疫情之初，百度研究院做到了快速響應，並在2020年4月正式對外發布了LinearDesign mRNA序列優化算法。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"與依靠窮舉法搜索穩定mRNA序列的思路不同，LinearDesign通過動態規劃算法，將序列穩定性和蛋白質翻譯效率指標進行聯合優化，可在10分鐘內找到比天然序列更加穩定且蛋白質表達水平更高的新冠mRNA疫苗序列，實現了在最短時間內用最高的效率得出最優化的方案。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"爲了進一步驗證LinearDesign算法的實際有效性，百度聯合mRNA藥物公司斯微生物進行了一系列實驗研究。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"本次實驗研究基於LinearDesign算法設計的七條編碼新冠病毒刺突蛋白的mRNA疫苗序列，對這些mRNA分子的穩定性、編碼蛋白表達水平以及動物免疫原性等方面開展了深入研究，並與用傳統方法設計的基準mRNA序列進行了頭對頭的對比。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"近日，百度和斯微生物聯合公佈了mRNA疫苗序列設計算法LinearDesign的新冠病毒疫苗生物實驗結果：\t在穩定性、蛋白質表達水平以及免疫原性等多個衡量疫苗的重要指標上，LinearDesign設計的新冠疫苗序列均大幅優於傳統方法設計的基準序列，尤其在抗原中和抗體滴度這一衡量疫苗有效性的關鍵指標上，LinearDesign設計的序列最高超過基準序列達到20倍，表示LinearDesign設計的疫苗能夠有效產生免疫保護；其實際有效性和應用於生物製藥領域的價值被充分驗證。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/4f\/ab\/4ff4a73dfc46abb3ab6a0bb9f349e8ab.png","alt":null,"title":"","style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":"","fromPaste":false,"pastePass":false}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","marks":[{"type":"size","attrs":{"size":10}}],"text":"百度LinearDesign算法設計的七條疫苗序列（A-G）以及基準序列（H）等相關信息"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"百度LinearDesign算法從理論層面和生物學實驗層面得到有效性驗證，爲將AI應用於生命科學探索出一條實際可行的道路。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"LinearDesign算法也進一步驗證了AI+製藥的價值。這項技術還具有廣泛的適用性，除了應用於新冠病毒mRNA疫苗的研發，還能包括傳染病疫苗、腫瘤疫苗、單抗等各種疫苗和藥物的研發，也進一步驗證了人工智能、生物計算技術在生命科學領域的實際應用價值。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"以LinearDesign算法爲代表的生物計算技術，將大大縮短新藥研發的週期、降低研發成本，並能提高醫療診斷的準確性和效率。"}]}]}