1 of 64

Computing and the Love of Words

Collaborations in Genomics and the Digital Humanities

1

Dr. Mark D. LeBlanc, Professor of Computer Science

Wheaton College, Norton, MA

cs.wheatoncollege.edu/mleblanc

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

2 of 64

2

made with BubbleViz

lexos3.wheatoncollege.edu

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

3 of 64

A Computational Linguist, Math Educator, and

a Cognitive Scientist walk into a bar …

3

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

4 of 64

A Computational Linguist, Math Educator, and a Cognitive Scientist walk into a bar …

“Traditional wording”

In the beginning, Alex had some marbles. Then Bethany gave Alex 3 more marbles. Now Alex has 8 altogether. How many marbles did Alex have in the beginning?

4

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

5 of 64

A Computational Linguist, Math Educator, and a Cognitive Scientist walk into a bar …

“Traditional wording”

In the beginning, Alex had some marbles. Then Bethany gave Alex 3 more marbles. Now Alex has 8 altogether. How many marbles did Alex have in the beginning?

Below: a more tersely-worded version sometimes found on exams.

Alex had some marbles. Bethany gave Alex 3 marbles. Now Alex has 8. How many marbles did Alex have?

5

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

6 of 64

6

1st and 3rd grader’s success rates on tests

LeBlanc, M.D. and Weber-Russell, S. (1996). Text integration and mathematical connections: a computer model of arithmetic word problem solving. Cognitive Science, v20(3), 357-407.

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

7 of 64

Me (by myself): My model “read” and solved word problems.

EDUCE (parser): 23,000 lines of LISP

SELAH (text integration/solver): 9,300 lines of LISP

7

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

8 of 64

Lesson #1ab

Me (by myself): My model “read” and solved word problems.

EDUCE (parser): 23,000 lines of LISP

SELAH (text integration/solver): 9,300 lines of LISP

Math Ed: “The wording matters to kids.”

Cognition: “Where are your stats? How does the model’s memory load and required text-based inferences compare with children’s success rates?

8

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

9 of 64

Lesson #1ab

Me (by myself): My model “read” and solved word problems.

EDUCE (parser): 23,000 lines of LISP

SELAH (text integration/solver): 9,300 lines of LISP

Math Ed: “The wording matters to kids.”

Cognition: “Where are your stats? How does the model’s memory load and required text-based inferences compare with children’s success rates?

9

(b) ground truth matters

(a) clients matter!

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

10 of 64

Working with BIO: The Genomics Research Group

circa 1998

10

Me (very naively):

Do you biologist have data?

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

11 of 64

Working with BIO: The Genomics Research Group

11

Science, Volume 282, 11 Dec 1998

First multi-cellular organism sequenced

1000 cells

Six Chromosomes

Entire Genome: 101 M base pairs

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

12 of 64

E.coli K12 genome (only 5M bp) on my Desktop

12

% ls -l Ecoli.fna 

-rw-r--r--+ 1 w00028110  523238136  4639298 Sep 14  2003 Ecoli.fna

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

13 of 64

E.coli K12 genome

>gi|170079663|ref|NC_010473.1| Escherichia coli str. K-12, complete genome

AGCTTTTCATTCTGACTGCAACGGGCAATATGTCTCTGTGTGGATTAAAAAAAGAGTGTCTGATAGCAGC

TTCTGAACTGGTTACCTGCCGTGAGTAAATTAAAATTTTATTGACTTAGGTCACTAAATACTTTAACCAA

TATAGGCATAGCGCACAGACAGATAAAAATTACAGAGTACACAACATCCATGAAACGCATTAGCACCACC

ATTACCACCACCATCACCATTACCACAGGTAACGGTGCGGGCTGACGCGTACAGGAAACACAGAAAAAAG

CCCGCACCTGACAGTGCGGGCTTTTTTTTTCGACCAAAGGTAACGAGGTAACAACCATGCGAGTGTTGAA

GTTCGGCGGTACATCAGTGGCAAATGCAGAACGTTTTCTGCGTGTTGCCGATATTCTGGAAAGCAATGCC

AGGCAGGGGCAGGTGGCCACCGTCCTCTCTGCCCCCGCCAAAATCACCAACCACCTGGTGGCGATGATTG

AAAAAACCATTAGCGGCCAGGATGCTTTACCCAATATCAGCGATGCCGAACGTATTTTTGCCGAACTTTT

GACGGGACTCGCCGCCGCCCAGCCGGGGTTCCCGCTGGCGCAATTGAAAACTTTCGTCGATCAGGAATTT

GCCCAAATAAAACATGTCCTGCATGGCATTAGTTTGTTGGGGCAGTGCCCGGATAGCATCAACGCTGCGC

TGATTTGCCGTGGCGAGAAAATGTCGATCGCCATTATGGCCGGCGTATTAGAAGCGCGCGGTCACAACGT

TACTGTTATCGATCCGGTCGAAAAACTGCTGGCAGTGGGGCATTACCTCGAATCTACCGTCGATATTGCT

GAGTCCACCCGCCGTATTGCGGCAAGCCGCATTCCGGCTGATCACATGGTGCTGATGGCAGGTTTCACCG

CCGGTAATGAAAAAGGCGAACTGGTGGTGCTTGGACGCAACGGTTCCGACTACTCTGCTGCGGTGCTGGC

TGCCTGTTTACGCGCCGATTGTTGCGAGATTTGGACGGACGTTGACGGGGTCTATACCTGCGACCCGCGT …

TATCGACTTACGTGTCTGCGGTGTTGCCAACTCGAAGGCTCTGCTCACCAATGTACATGGCCTTAATCTG

13

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

14 of 64

Lesson #2a

>gi|170079663|ref|NC_010473.1| Escherichia coli str. K-12, complete genome

AGCTTTTCATTCTGACTGCAACGGGCAATATGTCTCTGTGTGGATTAAAAAAAGAGTGTCTGATAGCAGC

TTCTGAACTGGTTACCTGCCGTGAGTAAATTAAAATTTTATTGACTTAGGTCACTAAATACTTTAACCAA

TATAGGCATAGCGCACAGACAGATAAAAATTACAGAGTACACAACATCCATGAAACGCATTAGCACCACC

ATTACCACCACCATCACCATTACCACAGGTAACGGTGCGGGCTGACGCGTACAGGAAACACAGAAAAAAG

CCCGCACCTGACAGTGCGGGCTTTTTTTTTCGACCAAAGGTAACGAGGTAACAACCATGCGAGTGTTGAA

GTTCGGCGGTACATCAGTGGCAAATGCAGAACGTTTTCTGCGTGTTGCCGATATTCTGGAAAGCAATGCC

AGGCAGGGGCAGGTGGCCACCGTCCTCTCTGCCCCCGCCAAAATCACCAACCACCTGGTGGCGATGATTG

AAAAAACCATTAGCGGCCAGGATGCTTTACCCAATATCAGCGATGCCGAACGTATTTTTGCCGAACTTTT

GACGGGACTCGCCGCCGCCCAGCCGGGGTTCCCGCTGGCGCAATTGAAAACTTTCGTCGATCAGGAATTT

GCCCAAATAAAACATGTCCTGCATGGCATTAGTTTGTTGGGGCAGTGCCCGGATAGCATCAACGCTGCGC

TGATTTGCCGTGGCGAGAAAATGTCGATCGCCATTATGGCCGGCGTATTAGAAGCGCGCGGTCACAACGT

TACTGTTATCGATCCGGTCGAAAAACTGCTGGCAGTGGGGCATTACCTCGAATCTACCGTCGATATTGCT

GAGTCCACCCGCCGTATTGCGGCAAGCCGCATTCCGGCTGATCACATGGTGCTGATGGCAGGTTTCACCG

CCGGTAATGAAAAAGGCGAACTGGTGGTGCTTGGACGCAACGGTTCCGACTACTCTGCTGCGGTGCTGGC

TGCCTGTTTACGCGCCGATTGTTGCGAGATTTGGACGGACGTTGACGGGGTCTATACCTGCGACCCGCGT …

TATCGACTTACGTGTCTGCGGTGTTGCCAACTCGAAGGCTCTGCTCACCAATGTACATGGCCTTAATCTG

14

(a) Computer scientists have hubris.

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

15 of 64

Lesson #2ab

>gi|170079663|ref|NC_010473.1| Escherichia coli str. K-12, complete genome

AGCTTTTCATTCTGACTGCAACGGGCAATATGTCTCTGTGTGGATTAAAAAAAGAGTGTCTGATAGCAGC

TTCTGAACTGGTTACCTGCCGTGAGTAAATTAAAATTTTATTGACTTAGGTCACTAAATACTTTAACCAA

TATAGGCATAGCGCACAGACAGATAAAAATTACAGAGTACACAACATCCATGAAACGCATTAGCACCACC

ATTACCACCACCATCACCATTACCACAGGTAACGGTGCGGGCTGACGCGTACAGGAAACACAGAAAAAAG

CCCGCACCTGACAGTGCGGGCTTTTTTTTTCGACCAAAGGTAACGAGGTAACAACCATGCGAGTGTTGAA

GTTCGGCGGTACATCAGTGGCAAATGCAGAACGTTTTCTGCGTGTTGCCGATATTCTGGAAAGCAATGCC

AGGCAGGGGCAGGTGGCCACCGTCCTCTCTGCCCCCGCCAAAATCACCAACCACCTGGTGGCGATGATTG

AAAAAACCATTAGCGGCCAGGATGCTTTACCCAATATCAGCGATGCCGAACGTATTTTTGCCGAACTTTT

GACGGGACTCGCCGCCGCCCAGCCGGGGTTCCCGCTGGCGCAATTGAAAACTTTCGTCGATCAGGAATTT

GCCCAAATAAAACATGTCCTGCATGGCATTAGTTTGTTGGGGCAGTGCCCGGATAGCATCAACGCTGCGC

TGATTTGCCGTGGCGAGAAAATGTCGATCGCCATTATGGCCGGCGTATTAGAAGCGCGCGGTCACAACGT

TACTGTTATCGATCCGGTCGAAAAACTGCTGGCAGTGGGGCATTACCTCGAATCTACCGTCGATATTGCT

GAGTCCACCCGCCGTATTGCGGCAAGCCGCATTCCGGCTGATCACATGGTGCTGATGGCAGGTTTCACCG

CCGGTAATGAAAAAGGCGAACTGGTGGTGCTTGGACGCAACGGTTCCGACTACTCTGCTGCGGTGCTGGC

TGCCTGTTTACGCGCCGATTGTTGCGAGATTTGGACGGACGTTGACGGGGTCTATACCTGCGACCCGCGT …

TATCGACTTACGTGTCTGCGGTGTTGCCAACTCGAAGGCTCTGCTCACCAATGTACATGGCCTTAATCTG

15

(a) Computer scientists have hubris.

(b) Be prepared to be WOWed!

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

16 of 64

Lesson #3: A good metaphor can take you a long way …

16

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

17 of 64

Lesson #3: A good metaphor can take you a long way …

Imagine that you have the text of

Alice in Wonderland ...

(but remove all the spacing, word boundaries, etc.)

17

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

18 of 64

Could we decipher Alice in Wonderland?

allrightsaidthecatandthistimeitvanishedquiteslowlybeginningwith theendofthetailandendingwiththegrinwhichremainedsometimeaftertherestofithadgone

  • (un)likely words
  • likely spelling rules
  • regular and

irregular grammar

18

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

19 of 64

Regulatory Motifs: On/Off switches for gene regulation

19

+

-

gene

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

20 of 64

Regulatory Motifs: On/Off switches for gene regulation

20

+

-

gene

Direct Repeats: ACGGacgg

Mirror Repeats: ACGTtgca

Inverted Repeats: ACGTacgt

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

21 of 64

Regulatory Motifs: On/Off switches for gene regulation

21

+

-

gene

Wait!?

Like palindromes?

Direct Repeats: ACGGacgg

Mirror Repeats: ACGTtgca

Inverted Repeats: ACGTacgt

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

22 of 64

Palindromes? Really? Like …

22

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

23 of 64

Palindromes? Really? Like …

23

Yup.

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

24 of 64

Lesson #4: Ask your prof, “Now what?”

24

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

25 of 64

25

Analysis by hand...

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

26 of 64

Analysis by hand...

26

We can help!

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

27 of 64

27

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

28 of 64

LeBlanc, M.D., Aspeslagh, G. ‘00, Buggia N. ‘01, and Dyer , B.D. (2000). An annotated catalogue of inverted repeats of Caenorhabditis elegans Chromosome III with observations concerning odd/even biases and conserved motifs, Genome Research, v10(9): 1381-1392.

28

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

29 of 64

29

Dyer, B.D., LeBlanc, M.D., Benz, S. ‘05, Cahalan, P.’04, Donorfio, B.’04, Sagui, P. ’04, Villa, A.’03, and Williams, G. ‘03 (2004). A DNA motif lexicon: cataloguing and annotating sequences. In Silico Biology, v4, 0039. 471-478.

30 of 64

30

Dyer, B.D., Kahn, M.J., and LeBlanc, M.D. (2007). Classification and regression tree (CART) analyses of genomic signatures reveal sets of tetramers that discriminate temperature optima of archaea and bacteria. Archaea 2:159–167.

Genomic signatures (counting “words”)

31 of 64

teaching together

31

COMP 215 Algorithms

BIO 211 Genetics

BIO/COMP 242 DNA

examples of

COMP 115 Python Programming

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

32 of 64

Course Materials (for other profs)

32

LeBlanc, M.D. and Dyer, B.D. (2007). Perl for Exploring DNA. Oxford University Press.

LeBlanc, M.D. and Dyer, B.D. (2004). Bioinformatics and Computing Curricula 2001 -- Why Computer Science is well positioned in a post-genomic world. ACM SIGCSE Bulletin, v36 (4), Dec. 2004, 64-67.

https://www.engage-csedu.org/

1_DNA: Playing with (strings of) DNA

2_DNA: Chargaff’s Numbers

3_DNA: Mini Gene Finder

4_DNA: Motif Finder

5_DNA: Comparative Genomics via Genomic Signature

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

33 of 64

Lesson #5: Got computational chops? You are needed more than ever!

33

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

34 of 64

Lesson #5: Got computational chops? You are needed more than ever!

today:

AlphaFold 3

Evo 2: a genome foundation AI model trained on over 9T DNA building blocks from 128,000 species

(the entire tree of life)

34

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

35 of 64

35

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

36 of 64

Lesson #1: Clients matter; ground truth matters.

Lesson #2: Be prepared to be WOWed!

Lesson #3: A good metaphor can take you a long way …

Lesson #4: Ask your prof, “Now what?”

Lesson #5: Got computational chops? You are needed more than ever!

36

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

37 of 64

Lesson #6: Be prepared to change course

37

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

38 of 64

Lesson #6: Be prepared to change course

and knowing you can help others makes you feel good

38

^

really

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

39 of 64

Lesson #6: Be prepared to change course

circa 2005

Mike Drout (English, Medieval, Tolkien scholar):

“Whoa! The entire (surviving) Anglo-Saxon (Old English) corpus

is on this CD-ROM!”

Me: “umm, cool.”

Drout: “Can you write programs to search for words?”

Me: (small sigh) “Yup. That’s pretty much what we do.”

39

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

40 of 64

Lexomics: A new close reading ….

Summers: 2008 - present

8-15 compSci and Humanities students

Wheaton internal funding and

40

lexomics.wheatoncollege.edu

lexos.wheatoncollege.edu

github — https://github.com/WheatonCS/Lexos

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

41 of 64

Lexomics: A new close reading ….

Summers: 2008 - present

8-15 compSci and Humanities students

Wheaton internal funding and

41

lexomics.wheatoncollege.edu

lexos.wheatoncollege.edu

github — https://github.com/WheatonCS/Lexos

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

42 of 64

From Whiteboard to Tools:

Corpus Construction with a Virtual Manuscript

42

Pick from the Poetry

Pick from the Prose

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

43 of 64

43

Lexos - a workflow for experiments with digitized texts

44 of 64

44

45 of 64

Beowulf: a very (very) close read …

45

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

46 of 64

Building Tools for a close read: Rolling Window

46

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

47 of 64

Feng, W. ’19 and LeBlanc, M.D. (2019). Top-10 Suggestions from a Decade of Managing Undergraduate Software Teams. The Journal of Computing Sciences in College, v34(6), 70-83.

Zhang, C. ’18, de Landaluce, A. ’17, Feng, W. ’19, Steffens, E. ’18, Kleinman, and LeBlanc (2018). Lexos 2017: Building Reliable Software in Python. The Journal of Computing Sciences in College, v33(6), 124-134.

47

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

48 of 64

Top-10 Suggestions from a Decade of Managing Undergraduate Software Teams

Feng, W. ’19 and LeBlanc, M.D. (2019). Journal of Computing Sciences in College, v34(6), 70-83.

2.1 Recruit actively and iteratively

2.2 Proceed the initial sprint with a bootcamp

2.3 Use continuous integration (CI) tools

2.4 Use a good, modularized structure

2.5 Apply unit testing

2.6 Conduct peer reviews

2.7 Meet in daily stand-ups

2.8 Lead toward success: assign easy tasks early

2.9 Improve one tool

2.10 Add new functionalities

48

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

49 of 64

Course Materials (for other profs)

49

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

50 of 64

Lesson #7: Embrace hard, open-ended problems

50

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

51 of 64

51

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

52 of 64

52

A Dream (1831) signed: “P”

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

53 of 64

53

LeBlanc, M.D. (2017). Toward Reproducibility in DH Experiments: A Case Study in Search of Edgar Allan Poe’s First Published Work. Digital Humanities 2017, Montreal, Canada, August 9, 2017 https://dh2017.adho.org/abstracts/027/027.pdf

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

54 of 64

54

Edgar

and brother Henry

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

55 of 64

55

Edgar

and brother Henry

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

56 of 64

56

Edgar?

and brother Henry

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

57 of 64

IF (stuck) goto Lesson #1: Clients matter

57

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

58 of 64

IF (stuck) goto Lesson #1: Clients matter

58

Ask a chatBot for help finding a domain expert?

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

59 of 64

IF (stuck) goto Lesson #1: Clients matter

59

omg! “Delvish” !!

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

60 of 64

Lesson #6: Be prepared to change course

Lesson #7: Embrace hard, open-ended problems

Lesson #1: Clients matter

60

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

61 of 64

Computing and the Love of Words

Collaborations in Genomics and the Digital Humanities

61

Dr. Mark D. LeBlanc, Professor of Computer Science

Wheaton College, Norton, MA

cs.wheatoncollege.edu/mleblanc

Computer Science with Purpose: Humanities, Life Sciences, Social Sciences - Feb. 25, 2025

thank you

62 of 64

do we have time to play with Lexos?

lexos.wheatoncollege.edu

62

63 of 64

63

64 of 64

64