¼¼¹Ì³ª/Çà»ç

> Á¤º¸¼¾ÅÍ > ¼¼¹Ì³ª/Çà»ç

Á¦¸ñ 2019³â 11¿ù 26ÀÏ »ý¹°Á¤º¸ÇÐ ¼¼¹Ì³ª
ÀÛ¼ºÀÚ °ü¸®ÀÚ ÀÛ¼ºÀÏ 2019-11-20

¼­¿ï´ëÇб³ »ý¹°Á¤º¸ÇÐ Çùµ¿°úÁ¤ ÁÖÃÖ·Î ¼¼¹Ì³ª¸¦ ¾Æ·¡¿Í °°ÀÌ ¿­°íÀÚ ÇÏ¿À´Ï ¸¹Àº Âü¿© ¹Ù¶ø´Ï´Ù.


ÀϽÃ:¡¡¡¡2019³â 11¿ù 26ÀÏ È­¿äÀÏ ¿ÀÀü 11½Ã

¿¬»ç:¡¡ ¼®Â÷¿Á ±³¼ö´Ô¡¡(¼­¿ï´ëÇб³)

Àå¼Ò:¡¡¡¡¼­¿ï´ëÇб³¡¡25-1µ¿ 1Ãþ (±¹Á¦È¸ÀǽÇ)


TITLE

What did DeepMind's AlphaFold contribute to the history of protein structure prediction?


ABSTRACT

¡¡The protein structure prediction problem, also called protein folding problem, remains one of the most difficult, unsolved problems in science. This problem has been tackled since the first protein structure was unraveled by x-ray crystallography in 1958. In December 2018, the Guardians reported that DeepMind¡¯s AlphaFold, a successor of AlphaGo, predicted 3D protein structures better than any other research groups in the international CASP protein structure prediction competition. Although AlphaFold did not appear out of nothing, it definitely inspired scientists including myself. In this seminar, I would like to share my own career history in protein structure prediction and to review how the protein structure prediction field evolved over the past CASPs including the advent of AlphaFold. If time allows, a view on how protein structure prediction can be applied to real world problems will be presented.