April week5   
*research(ai/bio/nlp-papers/lectrues/projects)  
** incremental methods .. 언제끝나냐       

*integrated(research and class)..only way I found?  
** robustness of nucl lms(and adv attack?)    

*class  
** RobustDeepLearning(assignment will release next next week)       
** ECE303TA    

* future...research    
** upload zenodo    
** iterative alignment algo  
** paper focus: E-NLP(??description?)    
** Aug focus(B)  
** deep hierarchy    
(future pro+nu)  
** generative benchmark  
(gf+rm+dc--->nucl)
(check nucl with blast--->check real nicl whether sim or not)  
** using drug bank web or db(good for using)  


## eng reading/listening/speaking/writing(paper)  
## math? code? (lecture/assignment/project)

Monday, April 29, 2024

research
** experiments
class
** study
o
주기적으로 ㄱㅃㅊㄱㅎㄴㅆㅂㄴㅇㅈㅉㄺㅆㅂㄴㅇ

Tuesday, April 30, 2024

research
** experiments
integrated
** ㄷㅎㅊ준비
class
** study
o
한글자음초성 handling not easy with lm(위의 예시)->좋은토픽

April week4   
*research(ai/bio/nlp-papers/lectrues/projects)  
** iBlast(G)     

*class  
** RobustDeepLearning(assignment will release this week)       
** ECE303TA    

* future...research    
** upload zenodo    
** paper focus: E-NLP(??description?)    
** Aug focus(B)  
** deep hierarchy    
(future pro+nu)  
** generative benchmark  
(gf+rm+dc--->nucl)
(check nucl with blast--->check real nicl whether sim or not)  
** using drug bank web or db(good for using)  


## eng reading/listening/speaking/writing(paper)  
## math? code? (lecture/assignment/project)

Monday, April 22, 2024

research
class
o

Tuesday, April 23, 2024

class
** writing assignment
ta
** meeting… 펑크내심..ㄷㄷㅂㄷㅂㄷ
research
** experiments
o

Wednesday, April 24, 2024

class
** writing assignment
research
** experiments
o

Thursday, April 25, 2024

research
** experiments
class
** writing assignment
o

Friday, April 26, 2024

research
** experiments
class
** code?
o

Saturday, April 27, 2024

research
** experiments
class
** study
o

Sunday, April 28, 2024

research
** experiments
class
** study
o
*** maybe for future, 적대적유전자입력 생성 및 공격.. 근데 이 유전자가 특정 생물임을 확인하는 절차(db에서 확인)가 있어야겠지…

-문제를 자기가 찾고 스스로 해결, 내가 찾은 문제가 제일 중요한 문제다라고 설명할 수 있어야!
역량 극단적으로 빠르게 발전..

-정확도가 제일 중요하긴하네… 성능(정확도)가 가장 보존되는 한도 내에서 속도나 다른 것을 업그레이드 시켜야!

-트랜스포머 뜯어보거나 nano gpt 뜯어보는거, 카파시가(openai 퇴사자) 어떻게 학습하는지 뜯어보면 좋아

April week3   
*research(ai/bio/nlp-papers/lectrues/projects)  
** iBlast(G)     

*class  
** RobustDeepLearning(assignment will release this week)       
** ECE303TA    

* future...research    
** upload zenodo    
** paper focus: E-NLP(??description?)    
** Aug focus(B)  
** deep hierarchy    
(future pro+nu)  
** generative benchmark  
(gf+rm+dc--->nucl)
(check nucl with blast--->check real nicl whether sim or not)  
** using drug bank web or db(good for using)  


## eng reading/listening/speaking/writing(paper)  
## math? code? (lecture/assignment/project)

Monday, April 15, 2024

research
** 23, 234, 2345, 23456, 234567, 2345678, 23456789, 2345678910 ok
** merge ok //and benchmark metric test ing
class
** summary of paper? or topics?
o

Tuesday, April 16, 2024

research
** benchmark metric test-accuracy etc.. ok
** Hit coverage ratio(non incremental) ok
** Protein Class Accuracy(non incremental) ok
** average evalues (calculate with best hit) draw plot ok
** number of total hit vs batches
*** average whole hits vs batches
class
** summary of paper? or topics?
o

Wednesday, April 17, 2024

research
class
** summary of paper? or topics?
o

Thursday, April 18, 2024

research
class
** summary of paper? or topics? ->must do today and think about 코워킹(넷트워킹)
*** 메일 보내야하나
o

Friday, April 19, 2024

research
class
** paper recommended
*** + 코워킹(넷트워킹)?
o

Saturday, April 20, 2024

research
class
** paper recommended
*** + 코워킹(넷트워킹)?
o

Sunday, April 21, 2024

research
class
** paper recommended
*** + 코워킹(넷트워킹)?
o

[slurm batch sh 파일 업데이트 ㅎㅎ] #!/bin/bash -l #SBATCH –partition def #SBATCH –account rosenmriprj #SBATCH –cpus-per-task 32 #SBATCH -t 00:10:00 #SBATCH –mem=2G #SBATCH -o slurm-%j-${2}.out # 표준 출력 파일 #SBATCH -e slurm-%j-${2}.err # 표준 오류 파일

conda activate myenv

nproc

echo “SLURM_CPUS_ON_NODE = $SLURM_CPUS_ON_NODE”

./../makeblastdb -in $1 -dbtype prot

#./../blastp -query astral-scopedom-seqres-gd-all-2.08-2023-01-06_subset_1.fa -db $1 -out $2 -outfmt 5 -num_threads 32

작업 완료 후 seff 명령어를 사용하여 작업 통계를 출력 파일에 추가

echo “SLURM job efficiency statistics:” » slurm-%j-${2}.out seff $SLURM_JOB_ID » slurm-%j-${2}.out [근데 의도대로 딱 파일명이 잘 찍히지는 않음.. 필요 정보는 찍긴함]

April week2   
*research(ai/bio/nlp-papers/lectrues/projects)  
** iBlast(G) ㅇㄱㄱㅇㄱㅃㄹㄲㄴㄱㅍㅇㅍㅃㄹㅈㅂㅎㅈ    
** paper focus: E-NLP  
  
** Aug focus(B)  
** deep hierarchy    
(future pro+nu)  
** generative benchmark  
(gf+rm+dc--->nucl)
(check nucl with blast--->check real nicl whether sim or not)  
** using drug bank web or db(good for using)  

*class  
** RobustDeepLearning     
** ECE303TA    
	
## eng reading/listening/speaking/writing(paper)  
## math? code? (lecture/assignment/project)

Monday, April 08, 2024

research
class
** summary of book
o

Tuesday, April 09, 2024

research
class
** summary of book
o

Wednesday, April 10, 2024

research
class
** summary of book
o

Thursday, April 11, 2024

research
class
** summary of book
o

Friday, April 12, 2024

research
class
** summary of book
o

Saturday, April 13, 2024

research
class
** summary of book
o

Sunday, April 14, 2024

research
class
** summary of book
o ㅅㅅㅂㅀㄱㅇㄴㅠㅠ

makeblastdb -in source.fasta -dbtype prot sed -i 's/\x13//g' astral-scopedom-seqres-gd-all-2.08-2023-01-06_subset_1.fa singularity exec --bind /ifs/groups/rosenMRIGrp/hty23/astral_scope/:/data /ifs/groups/rosenMRIGrp/hty23/iMethods/methods/blast_latest.sif makeblastdb -in /data/astral-scopedom-seqres-gd-all-2.08-2023-01-06_subset_2.fa -dbtype prot singularity exec --bind /ifs/groups/rosenMRIGrp/hty23/astral_scope/:/data /ifs/groups/rosenMRIGrp/hty23/iMethods/methods/blast_latest.sif blastp -query /data/astral-scopedom-seqres-gd-all-2.08-2023-01-06_subset_1.fa -db /data/astral-scopedom-seqres-gd-all-2.08-2023-01-06_subset_2 -out /data/result/blastp_astral_scope_result2.xml -outfmt 5 -----음 근데 signularity라 그런가 잘 안됨... 그냥 직접 다운로드 후 실행해야할듯 wget https://ftp.ncbi.nlm.nih.gov/blast/executables/blast+/2.12.0/ncbi-blast-2.12.0+-x64-linux.tar.gz tar -xzvf ncbi-blast-2.12.0+-x64-linux.tar.gz bin에 실행파일있음 blastp, blastn 등등.. makeblastdb -in /content/drive/MyDrive/playground_test_anything/astral-scopedom-seqres-gd-all-2.08-2023-01-06_subset_2.fa -dbtype prot 아 근데 sed -i 's/\x13//g' /content/drive/MyDrive/playground_test_anything/astral-scopedom-seqres-gd-all-2.08-2023-01-06_subset_2.fa 이거해줘야.. (search:) ./../blastp -query astral-scopedom-seqres-gd-all-2.08-2023-01-06_subset_1.fa -db astral-scopedom-seqres-gd-all-2.08-2023-01-06_subset_2.fa -out blastp_astral_scope_result2.xml -outfmt 5 -num_threads 96 ./../blastp -query astral-scopedom-seqres-gd-all-2.08-2023-01-06_subset_1.fa -db astral-scopedom-seqres-gd-all-2.08-2023-01-06_subset_2.fa -out blastp_astral_scope_result2.m8 -outfmt 6 -num_threads 96 get mmseqs github repo : git clone https://github.com/soedinglab/MMseqs2.git cd MMseqs2 mkdir build cd build module load cmake cmake .. make -j $(nproc) (search:) mmseqs easy-search query.fasta targetDB.fasta results.tsv tmp --threads 96 get diamond github repo : git clone https://github.com/bbuchfink/diamond.git cd diamond mkdir build cd build module load cmake cmake .. make -j $(nproc) (makedb:) ./../diamond makedb --in reference.fasta -d dbname (search:) ./../diamond blastp -d reference -q queries.fasta -o matches.tsv ./../diamond blastp -q astral-scopedom-seqres-gd-all-2.08-2023-01-06_subset_1.fa -d db_diamond_astral_scope_3 -o diamond_astral_scope_result3.m8 ./../diamond blastp -q astral-scopedom-seqres-gd-all-2.08-2023-01-06_subset_1.fa -d db_diamond_astral_scope_6 -o diamond_astral_scope_result6.xml --outfmt 5 [next? 23, 234, 2345, 23456, 234567, 2345678, 23456789, 2345678910]

April week1   
*research(ai/bio/nlp-papers/lectrues/projects)  
** iBlast(G) ㅇㄱㄱㅇㄱㅃㄹㄲㄴㄱㅍㅇㅍㅃㄹㅈㅂㅎㅈ    
** paper focus: E-NLP  
  
** Aug focus(B)  
** deep hierarchy    
(future pro+nu)  
** generative benchmark  
(gf+rm+dc--->nucl)
(check nucl with blast--->check real nicl whether sim or not)  
** using drug bank web or db(good for using)  

*class  
** RobustDeepLearning     
** ECE303TA    
	
## eng reading/listening/speaking/writing(paper)  
## math? code? (lecture/assignment/project)

Monday, April 01, 2024 * research * class o ** ㅍㅇㅍletsgo Tuesday, April 02, 2024 * research ** ㅎㄷㅂㅅㄱㅇㄱ * class ** summary of book o ** ㅍㅇㅍletsgo Wednesday, April 03, 2024 * research * class ** summary of book o ** ㅍㅇㅍletsgo Thursday, April 04, 2024 * research * class ** summary of book o ** reasonable advice is good Friday, April 05, 2024 * research * class ** summary of book o ** having good class with nice one Saturday, April 06, 2024 * research * class ** summary of book o ** ㄱㄴㄱㅃㅊㄴ Sunday, April 07, 2024 * research * class ** summary of book o